Stanford-Harvard research translates California test scores into months of missed learning

Credit: iStock_grady reese

A large-scale data project by researchers at Harvard and Stanford universities, released Friday, converted national and state test scores into equivalent rates of learning declines during the pandemic for every school district and student group in California and 28 other states.

Interactive data in Education Renewal Scorecard to clearly and rigorously confirm differences in performance and learning loss among states, districts within states, and racial, ethnic, and income groups within districts for grades three through eight. (Start here with California and other states.)

For California, the math data translated into a straight downward line on the graph, showing that wealthier districts, those with small percentages of low-income students, saw the least learning loss — in Orinda, Lafayette, a matter of weeks. and Los Gatos in the Bay Area—compared to the loss of three-quarters or more in grade-equivalent grades in the Bakersfield City and Mountain View Elementary School districts. (Go here for an interactive graphic showing the decline in math learning for California counties – blue circles.)

“Every poor district has lost a lot and rich districts have lost little, and that’s a picture that leaders in Sacramento should be responding to,” said Sean Reardon, a professor at the Stanford Graduate School of Education and senior fellow at the Stanford Institute for Economic Policy. The research that guided the data analysis.

The data also shows that states that were largely open to in-person contact during the height of the pandemic in 2020-21 did not fare significantly better than those whose students were learning remotely for most of that time. Students in every state dropped in math compared to where they were before Covid, but the drop in California, the last major state to fully reopen, was at the equivalent of 48% of grade level. That was the same as Florida and slightly better, with a 57% loss, than Texas, two states where schools remained open all year. The 12% decline in reading was nearly the same for all three states.

However, California’s loss of nearly half a year of learning in math, added to prior years, still placed it nearly a year behind the district’s average math score nationally.

By comparing the learning loss to their share of the $190 billion in federal Covid aid that states and districts received, the researchers calculated how far the funding could close the gap. Their conclusion: even strategies such as tutoring or more time to study are not enough. In two-thirds of districts nationwide, the grade-equivalent equivalent of learning losses exceeded what the funding they receive from the US bailout plan can likely provide. It’s the largest source of federal money to address learning loss, and it runs out in two years.

That’s critical news for parents because surveys suggest most parents believe students have already created almost everything they haven’t learned, said Thomas Kane, professor and faculty director of the Center for Education Policy Research at Harvard. who co-led the project. “There is a disconnect. Parents think that students are 95% recovered, but the average loss of a pandemic is half a year. So no leader can propose unpopular things like extending the school day and year if parents think everything is fine.”

For example, San Jose’s Alum Rock Union School District lost the equivalent of 84% of its grades during the pandemic, but the U.S. bailout provided enough to cover only 23% of the district’s school budget. More than three times that amount would be needed to make up for missed learning, assuming effective learning strategies were targeted.

But Stanford’s calculations don’t include additional state funding, which in California’s case is substantial: $11 billion, including $4 billion to expand after-school and summer education for low-income districts, on top of this year’s $15 billion in the U.S. rescue plan. An additional $4 billion has been earmarked for each over the next few years to expand mental health services and establish community schools, where the extra money can be used to hire counselors and pay for social services as well as academics.

Source: Education Recovery Scorecard by research centers at Harvard and Stanford universities

All of these selected California counties fell behind in math between 2018-19 and 2021-22, although the equivalent drop in literacy was smaller for affluent counties like the San Ramon Valley, as measured by the percentage of families eligible for free and reduced-price meals ( FRPL). Bakersfield City’s decline was almost the equivalent of a learning year. The national average was 48%.

Reardon said funding is also important. “The drop in scores is a symptom of the pandemic’s damage, but many other features of the pandemic have affected children’s lives. If we had mental health and well-being measures, we would probably see significant falls across all areas,” he said. “It’s important to think of harm as more than just a school problem.”

Kane and Reardon used the same national data from National Assessment of Educational Progress, or NAEP and Smarter balanced results for California from the 2018-19 and 2021-22 pre-pandemic periods that EdSource reported this week. However, researchers generally cannot compare test scores across states. Reardon’s team at Stanford overcame this challenge by creating a scale that aligned annual state test scores with fourth- and eighth-grade NAEP scores. NAEP served “as a kind of Rosetta plate” to measure district-level changes in academic skills, he said.

Scoreboard graphics and data identify high and low performers among California districts. Fremont Unified in Alameda County, where a third of its students are low-income, lost only a tenth of a year in math learning and gained 16% in reading compared to the national average.

Garden Grove Unified in Orange County, where two-thirds of students come from low-income families, dropped just 32% in math over the year, well above the state and national averages, and was flat in reading. Monterey Peninsula Unified was another better performer among low-income districts.

Nationally, the decline in success rates varied widely among districts within the same state. “The pandemic was like a bunch of tornadoes that swept across the country. Some communities were left relatively untouched while neighboring schools were devastated,” Kane said.

While the middle school district lost the equivalent of 52% in math and 23% of expected annual growth in reading, 2.5% of students were in districts where math scores increased and 5.3% of students were in districts where scores dropped by more than one. degree. The latter include Oklahoma City, New Haven, Connecticut and, with a two-year loss, Richmond, Virginia.

In reading, 14.8% of students were in districts that improved, and 1.4% were in districts that lost more than one grade.

Scorecard covers 4,000 school districts and 12 million students in 29 states that have reported test data. More states will be added as states release their scores, Reardon said.

For more news like this, click here to sign up for EdSource’s free daily email on the latest developments in education.


Leave a Comment