Insights, discoveries, and analysis from Harvard scholars and scientists.
A new report from the Education Recovery Scorecard, a collaboration between the Center for Education Policy Research at Harvard and the Educational Opportunity Project at Stanford University, shows that academic achievement gaps that widened during the pandemic have worsened in some states, including Massachusetts.
The study examined math and reading test scores in grades 3-8 in approximately 8,000 school districts in 30 states from spring 2019 to spring of last year. In a conversation about the findings, co-author Thomas Kane urged school districts to invest remaining pandemic aid on academic recovery efforts before the funds expire in the fall. The interview was edited for clarity and length.
From your report, it seems the pace of recovery has been uneven. Could you give an overview of your findings?
If you looked across all the states, the recovery last year was actually large by historical standards. The recovery was twice as large as the average annual rate of change on the National Assessment of Educational Progress from 1990 to 2019 and 50 percent larger than the annual rate of change from 1990 to 2013, when math scores grew most rapidly. So, it was large, but it varied by state. Some states saw much bigger increases than others. But the most troubling finding was that higher-poverty districts which lost the most during the pandemic did not close the gap nationally. In some states, like Massachusetts, those gaps grew between 2022 and 2023.
There were some bright spots. Alabama in math, and Louisiana, Illinois, and Mississippi in reading. Can you explain?
Alabama is the only state to be back above its pre-pandemic achievement in math, and there are three states that are above their pre-pandemic achievement in reading: Illinois, Mississippi, and Louisiana. But this doesn’t mean that Alabama is finished with its recovery. Even in Alabama, the students in Montgomery are still about half a grade level behind. So, yes, there has been progress nationally and, in a few states, their average achievement is back above 2019 levels. But in most states, the achievement gaps between the high-poverty and the low-poverty districts are wider than they were in 2019.
In the states that have recovered significantly, are there lessons for other places?
At this point, we can’t say exactly what made Alabama different. Our report is analogous to the National Assessment of Educational Progress — we are describing where progress is being made, but we’re not yet evaluating the efficacy of policies. There will be no National Assessment of Educational Progress for 2023, and so what we’ve done is we’ve used the state test scores to provide an alternative during this critical year, before the federal dollars run out. We’re describing what happened to achievement, not just at the state level, but at the individual district level. Soon, we and others will be using these data to understand what distinguished Alabama from Massachusetts. We’re just trying to get these [findings] out to inform policymakers and school districts while there’s still time.It’s especially important right now because there are only eight months left before the federal pandemic relief dollars expire.
Achievement gaps are wider in some places than others. How so?
As we reported last year, students in high-poverty school districts lost more ground than students in higher-income ones during the pandemic. The gaps that were already there in 2019 widened during the pandemic. The most important message from this report is that those widened gaps have not closed. In fact, in some states, just the opposite happened. Even though they lost less ground during the pandemic, wealthier districts like Newton, Wellesley, and Arlington began to recover between 2022 and 2023 while districts like Fall River, Lynn, and Revere, which have high proportions of students experiencing poverty, lost additional ground between 2022 and 2023.
There were some states where the recovery is being led by the poorer districts, but even there — even in Alabama, where the recovery was larger for districts like Birmingham — the recovery wasn’t enough to completely eliminate the increase in inequality that occurred during the pandemic. So even in Alabama, the poorer districts are lagging further behind their own 2019 achievement as the higher-income districts now exceed their 2019 achievement.
You mention two specific challenges for school leaders following the pandemic: teacher shortages and an increase in student absenteeism. How significant are these concerns in the recovery story?
We haven’t quantified the role of absenteeism in the recovery, but we know from research that each day a student is absent results in lost learning, and, when many students are returning from absences, it disrupts learning for other students in the classroom because a teacher is constantly having to reteach topics. Future research will have to show just how big a role student absenteeism played, but, while we’re waiting for that research, communities ought to be doing whatever they can to try to lower absenteeism rates. Absenteeism is one of the very few things that organizations outside of schools can help schools improve. Most mayors can’t teach Algebra 1, but they can do a public information campaign or provide public transportation passes to students to try to lower absenteeism. One of our mistakes as a country, I think, has been to see the recovery as primarily the job of schools. And of course, schools will have to be doing the classroom instruction, but mayors’ offices and churches and other community organizations ought to be looking for ways they can help. Reducing absenteeism is one of the clearest examples of those.
What other advice do you have to make up for interrupted learning?
I think one place to start is by letting parents know when their child is behind. A number of polls have reported that parents believe that their own children have already caught up. They have been misinformed. Parents could play a role in advocating more spending on academic recovery. Districts will spend the money on something; few of these dollars are going to be returned to the federal government. The point now is to get districts to spend the remaining funds to extend the recovery into next year.
Should there have been more requirements that federal aid be used on school recovery efforts? In your report, you say that districts were only required to spend 20 percent of the money they received on academic recovery. What else has the money been spent on?
When the American Rescue Plan passed in March 2021, no one knew how bad the losses would be. We knew remote learning was not the same as in-person, but many were hoping that hybrid learning may have been 75 or 80 percent as effective. It was not. Unfortunately, Congress only required districts to spend 20 percent on academic recovery, and 90 percent of the K–12 aid was sent directly to districts — leaving federal and state agencies with no leverage for coordinating recovery efforts. Imagine if, instead of launching a massive effort to develop a vaccine, the federal government had just handed money to local public health departments to find their own treatments for COVID. That’s what we did with the academic recovery. There was no coordination, little sharing of resources. Although some districts spent more on academic recovery, many more spent the money on salary increases, HVAC systems, new curricula, additional support staff working in schools. Thirteen thousand-plus school districts out there have been inventing their own recovery plans. Maybe we shouldn’t be that surprised that some districts have figured it out, but a lot of districts haven’t, and the recovery has really varied by district and by state.