The online tool, its creators say, will democratize school performance data at an important time, as schools nationwide struggle to recover from the COVID-19 pandemic.

Scheduled to go live today, the new website sports a simple interface that allows users to query it conversationally, as they would a search engine or AI chatbot, to plumb math and English language arts data in grades 3-8. At the moment, there are no firm plans to add high school-level data.

If, for instance, a superintendent is curious about math scores for kids learning English in her state, she can : 鈥淪how me math scores over time for English learners and non-English learners in Minnesota.鈥� Want to know the 10 school districts in Mississippi with the highest ELA scores in 2023? .

Similarly, parents moving to a new town or neighborhood can ask about data for individual schools in most cases.

The project, dubbed , is a partnership between Brown University and , the company that built the site鈥檚 AI functionality. 

The tool takes a cue from data dashboards, such as the federal government鈥檚 , which collect statewide assessment information. This one goes further, allowing more up-to-date analyses of state, district and even school-level data, with protections that shield individual students鈥� scores in small districts and schools. 

Within state, local and school-level data, users can also break down results by race, ethnicity, economic level and other indicators.

The AI aspect allows users to query the database in plain language, said Emily Oster, a well-known economist who often writes on parenting. Oster led the tool鈥檚 development and said its potential customer base is broad, from parents and school board members to state policymakers and journalists.

鈥淵ou can imagine people actually wanting to see in a more granular way, or be able to explore in a more granular way: 鈥楬ow are different schools in this district doing鈥� or 鈥楬ow is my district doing relative to another district?鈥� This will make that much easier.鈥�

Oster said the tool is so easy to use that a school board member sitting in a board meeting could pull out her phone and in a few seconds produce a chart showing school-by-school test results districtwide. 

Policymakers could also benefit from the tool, she said, since they can鈥檛 always access state assessment data without cumbersome requests to state education officials. 鈥淎nd that takes time. If you want to have access to get an insight quickly, this is going to make it easier.鈥�

What鈥檚 perhaps most useful, Oster said, is the ability to look inside individual states, down to the district or school level, to figure out which schools and populations are doing better than others. 鈥淚 think that’s actually pretty powerful in terms of where the policy is made.鈥�

Reliance on 鈥榩lain language鈥�

Project Manager Clare Halloran said Zelma grew out of Brown researchers鈥� own frustration in trying to compare COVID recovery data across states. 鈥淚t was usually hard to find out where the information was, what was missing,鈥� she said.

Even states with public-facing data portals and dashboards don鈥檛 make the job easy, she said, as many are 鈥渁 little bit clunky.鈥� They rely on dropdown menus that can only offer one indicator at a time. With Zelma, she said, 鈥淵ou can really just kind of say in plain language what you’re looking for,鈥� even if it involves several variables. 

鈥淚 think it will make a lot of data just a lot more accessible to a lot more people,鈥� she said. 鈥淲hen the states release their data, we get the headline. But it’s hard for the average person to explore it a little bit more.鈥�

All queries are public but the authors aren鈥檛 identified. The site resembles a Twitter-like feed, with the most recent queries at the top so users can see what others want to learn about.

It also offers warnings 鈥� dubbed 鈥渘otable events鈥� 鈥� that caution users not to read too much into proficiency levels in certain cases, such as in states and districts where new assessments are being administered, or where they see lower participation rates.

And while it can offer rudimentary comparisons between states, Oster said neither Zelma nor the assessments themselves are built for such comparisons. 

鈥淭here are things across states you might get out of this, for example how much recovery has there been鈥� in one state vs. another, she said. 鈥淵ou can sort of squint a little and think about differences in trends. And I actually think there is some stuff we can learn from those kinds of trends. But in terms of levels, these data are just not well suited to a question of, 鈥業s Mississippi outperforming Michigan?鈥� That’s why we’ve got the NAEP data.鈥�

Actually, asking the tool to compare states will prompt a warning saying that states administer different assessments and that proficiency rates 鈥渁re not comparable across states.鈥� 

Even with a more user-friendly interface, though, the site is only as good as the data underlying it 鈥� and it鈥檚 uneven among states. Minnesota, for instance, offers test scores clear back to the late 1990s. But Rhode Island has no data before 2018.

And, of course, virtually no states returned test scores in 2020 and 2021, when the U.S. Department of Education granted blanket standardized testing waivers amid the pandemic.

Paul Peterson, who directs the Program on Education Policy and Governance at Harvard University鈥檚 Kennedy School, said he welcomed the ability to more easily dig into states鈥� updated testing data.

鈥淎ny enhancement of transparency is a good thing,鈥� he said.

Disclosure: Walton Family Foundation provides financial support to Zelma and 蜜桃影视.

Then the pandemic hit and 鈥渢here were bigger fish to fry,鈥� said Principal Beth Farmer.

But the state had a plan.

Teachers in those schools would receive two years of training in what鈥檚 known as the science of reading and use a new curriculum with explicit phonics instruction. Farmer has already seen the payoff: Seventy-five percent of third graders met the goal this year, with similar improvement in fourth and fifth grades.

鈥淲hat appeared to be some penalty 鈥� has ended up being a gift,鈥� she said.

The progress sunk in when she recently talked to a student after a quiz. 鈥淪he said, 鈥業 was reading 14 words per minute, and now I can read 43 words per minute.鈥� When a kid can verbalize that to you, that’s real impact.鈥�

Greenville, with roughly 77,000 students, is South Carolina鈥檚 largest district, so the results figure significantly into the state鈥檚 overall average. Fifty-four percent of third through eighth graders statewide scored proficient or above this year in English language arts 鈥� a big jump from the 45% of students at that level in 2019.

While most states remain behind, South Carolina and three others 鈥� Iowa, Mississippi and Tennessee 鈥� are recovering from or exceeding COVID-related declines in reading, according to researchers at Brown University. Iowa and Mississippi have also surpassed their 2019 performance in math. Experts say improvements in literacy instruction and an accelerated return to in-person learning are among the key policy decisions contributing to the rebound.

鈥淚 am encouraged to see some states surpassing 2019,鈥� said economist Emily Oster, who leads Brown鈥檚 鈥淭his suggests substantial recovery is possible, and it provides an opportunity for learning.鈥�

She said it鈥檚 鈥渃rucial鈥� to understand what those states have done right.

In Iowa, more than 80% of schools offered in-person learning during the 2020-21 school year, according to state officials. In January of 2021, Gov. Kim Reynolds signed a law mandating that schools offer families in-person learning five days a week.

That鈥檚 likely one reason why the achievement declines in Iowa were not as steep as those in other states, said Heather Doe, a spokeswoman for the Iowa Department of Education. Between 2019 and 2021, the proficiency rate in English language arts dropped just 2 percentage points, compared with at least twice that much in several other states. 

Once more state results are released, Oster plans to match the data with the length of time schools were closed during the 2020-21 school year, as she did last year. The from the previous report was that states where schools were closed longer saw bigger drops in proficiency 鈥� as high as 20%.

In the other states, leaders overhauled the way students learn to read, a shift that is now showing up in test results. 

and were among the first states to adopt reform efforts that included a strong emphasis on foundational reading skills.

The turnaround in Mississippi 鈥� which in 2019 saw a dramatic leap in fourth grade reading scores on the National Assessment of Educational Progress 鈥� has garnered much attention and analysis. But a similar push was underway in the Palmetto State.

The state assigned reading specialists to schools that needed to improve, like Westcliffe. And it gave districts a list of recommended curricula. Greenville chose a program from , which Jeff McCoy, the district鈥檚 associate superintendent for academics, described as more 鈥渟cripted鈥� than the district鈥檚 prior approach.

鈥淲e recognized that phonics was a missing component,鈥� he said.

The 2023-24 passed this year included $39 million to make a highly regarded training course 鈥� Language Essentials for Teachers of Reading and Spelling 鈥� available to all K-3 teachers. 

is a more recent addition to the states requiring training and curriculum on foundational reading skills. Its literacy law passed in 2020. The state also used relief funds for and high-dosage tutoring.

Dale Chu, a fellow with the conservative Thomas B. Fordham Institute and a policy consultant who focuses on assessment, sees an additional reason for achievement gains in Tennessee: Despite the pandemic, the state was less divided over education.

鈥淯nlike any other state, they鈥檝e largely had bipartisan continuity on education policy across three administrations,鈥� he said.

Parent advocate Sonya Thomas, executive director of Nashville Propel, said the scores are good news for students in the early grades since Tennessee 鈥渟pent several decades鈥� in most educational rankings. But she鈥檚 less optimistic about older students鈥� opportunities to catch up. Many, she said, are 鈥渟everal grade levels behind.鈥�

鈥楪ive this some time鈥�

Despite the positive developments, researchers and testing experts urge caution about interpreting the increase in proficiency rates as a sign of true recovery. 

Scott Marion, president and executive director of the Center for Assessment, said Oster provides 鈥渁 pretty useful look鈥� at where states stand. But assessments aren鈥檛 comparable across states; what counts as proficient in one isn鈥檛 necessarily the same in another. 

Overall proficiency rates also tell just part of the story. In South Carolina, for example, racial achievement gaps haven鈥檛 changed much. In 2018, there was a 45 percentage point difference in proficiency rates between Asian and Black students in English language arts. Now it鈥檚 43 points. In math, the gap has actually increased 鈥� from 52 to 54 percentage points.

Additionally, some students never cross the threshold from one achievement category to the next, in terms of going from 鈥渄oes not meet expectations鈥� to 鈥渁pproaches expectations,鈥� said Dan Goldhaber, director of the Center for Analysis of Longitudinal Data in Education Research at the American Institutes for Research.

鈥淚鈥檓 particularly worried about kids at the bottom, who were unlikely to be proficient before or after the pandemic,鈥� he said.

In most states, proficiency rates in reading are still stuck below pre-pandemic levels. Scores in math are headed in the right direction; nearly all are 鈥渕aking progress,鈥� according to Oster. 

But her summary serves as a reminder of how long it will take for some students to rebound, Chu said. 鈥淚f you look at learning loss and what schools need to do to catch up, there鈥檚 no precedent,鈥� he said. 鈥淭he [education] system has never done this before.鈥�

Despite billions in federal relief funds for tutoring, summer school and extra staff in the classroom, five states 鈥� Arkansas, Delaware, Massachusetts, Minnesota and Nevada 鈥� have continued to lose ground in reading since 2019. The percentage of students scoring proficient or above dropped this year.

Minnesota, for one, is several years behind states like Mississippi in requiring reading instruction to include phonics. Just this year, the state passed the , legislation that provides $70 million for 鈥渟cience of reading鈥� training and curriculum. 

Last month, the Minneapolis district鈥檚 disappointing literacy results sparked a at a school board meeting. 

鈥淚 would not say that it is a privilege to share this data,鈥� Sarah Hunter, the district鈥檚 executive director of strategic initiatives, told the board. Since 2022, the percentage of district students who scored proficient decreased from 42% to 41% 鈥� the third consecutive year of decline. Board members blamed the pandemic and urged patience.

鈥淚 know our scores are still low,鈥� said . 鈥淟et’s give this some time.鈥� 

 Such comments left some advocates feeling uneasy.

鈥淗ow do we hold districts accountable?鈥� asked Josh Crosson, executive director of EdAllies in Minneapolis. 鈥淲e have a lot of funding that goes to schools that aren鈥檛 doing well in literacy.鈥�

He thinks the READ Act is a step forward, but doesn鈥檛 do enough to integrate literacy training and teacher preparation. 

鈥淚 don鈥檛 think we’re going to see improved outcomes in these first couple months,鈥� he said. 鈥淚 think we鈥檙e going to see improved outcomes in the next few years.鈥�

Districts that spent the majority of that year learning remotely tended to lose more ground in fourth-grade math scores than districts that reopened sooner. Every 10 additional days of school closures was associated with a roughly 0.2-point loss on NAEP from 2019 to 2022. The pattern was statistically significant and held even when controlling for the share of students eligible for free- or reduced-price lunch, a proxy for poverty.

鈥淭he districts with more remote learning have larger test score losses,鈥� said Emily Oster, a Brown University economics professor who has tracked school closures through the pandemic. 

鈥淚t鈥檚 pretty consistent with what we have seen up until now,鈥� added the researcher, an early and ardent supporter of reopening schools during the pandemic shutdown whose positions were .

The finding adds to the that online learning during the pandemic had a negative impact on student learning outcomes, even while there is renewed debate over how strongly the 2022 NAEP scores reflect it. The highly anticipated results released Monday showed the largest drops ever recorded in 4th and 8th grade math.

Peggy Carr, head of the U.S. Department of Education center that administers NAEP exams, played down any possible relationships between school closures and test results.

鈥淭here is nothing in this data that tells us there is a measurable difference between states and districts based solely on how long schools were closed,鈥� she said during a Friday press conference.

Oster, who also of the relationship between remote learning and NAEP results, called the National Center for Education Statistics director鈥檚 statement 鈥渙dd鈥� and 鈥渘ot very consistent with what we are seeing in the data.鈥�

However, she acknowledged that there is an element of truth to Carr鈥檚 words.

鈥淢aybe what they’re saying is that [school closure] is not the only determinant, and that鈥檚 right. It is not the case that there is a straight line between remoteness and test score losses,鈥� she said.

An NCES spokesperson affirmed that stance Tuesday, denying any 鈥渟imple direct relationship between duration of remote learning and score declines based on NAEP results鈥� in a statement emailed to 蜜桃影视.

鈥淐ontrolling for free- or reduced-price lunch is helpful but not sufficient,鈥� the spokesperson continued. 鈥淣CES will be conducting analyses that conform to the highest statistical standards, consider multiple variables and link data collected by NCES to other high quality datasets.鈥�

On the whole, results from what鈥檚 known as the Nation鈥檚 Report Card revealed the stark drop offs in math and a slide in reading since 2019, the last time the exam was administered. Some individual school systems, however, performed better than expected, including Los Angeles, among the districts which stayed in remote learning the longest and which saw improvements in reading for fourth graders and in both reading and math for eighth graders.

Since the release of NAEP results on Monday, and have conducted several analyses correlating scores with length of school closures and found moderate, statistically significant links. However, those analyses have largely focused on state data, an approach some experts warn against because it lumps districts that reopened quickly with those that stayed shuttered much longer.

鈥淲ithin states, there’s a lot of heterogeneity in terms of closure policies,鈥� said Tom Loveless, a longtime education researcher who formerly led the Brookings Institution鈥檚 Brown Center on Education Policy.

鈥淟ooking at district data is superior to looking at state data because that’s where the [reopening] decisions were made,鈥� he said.

蜜桃影视 took the district-level approach, crunching data from a sample of large urban school systems included in the NAEP release. Their scores were then matched with closure data from Oster鈥檚 , which tracked the percentage of the 2020-21 school year that districts offered remote, hybrid or in-person instruction. From the full sample of 26 school systems, Fresno was removed because it had no publicized 2022 NAEP scores and New York City, the nation鈥檚 largest school district, and Shelby County, Tennessee were excluded because they had no district-level school closure data available in the Hub.

Among the 23 remaining school systems, fourth-grade math was the only subject with a statistically significant relationship between district performance and time spent in remote learning. There were weak correlations in fourth-grade reading and eighth-grade math and no association for eighth-grade reading.

鈥淚t was very hard for the little kids to focus on Zoom,鈥� said Robin Lake, director of the Center on Reinventing Public Education. 鈥淚t wouldn’t surprise me if the younger students saw more of an impact on literacy skills and early foundational computational skills.鈥�

Her research group analyzed data on the effects of school closures, finding , especially for younger students and those living in poverty. 

鈥淪chools stayed closed too long, especially in urban areas,鈥� Lake said, noting that her judgment is much easier to make now with the benefit of hindsight as opposed to during the height of COVID when the science on infections and transmissibility was still coming into focus.

The variation in the NAEP results represents 鈥渟hades of badness,鈥� she said. 鈥淪ome states are celebrating not being as bad as other states, but nobody has much to celebrate here.鈥�

NAEP results must be interpreted carefully, experts caution. They are built to show how students are doing, not to explain the reasons behind their performance, Loveless said. (He compared the exam to a thermometer: 鈥淚t can tell you if you have a temperature, but it can鈥檛 tell you why.鈥�)

However, the exam is also the only U.S. test administered to students in all 50 states, making it 鈥渢he only game in town when it comes to comparing across states,鈥� said the former Brookings Institution researcher.

蜜桃影视 analysis, he said, 鈥渕akes an addition鈥� to the continued dialogue on the impacts of school closures during the pandemic.

Now, with the extent of pandemic missed learning coming into greater focus across the nation, Lake said, it鈥檚 time to hone in on how to respond.

鈥淲e’ve just got a lot of work to do to give kids back what they were owed, both academically and developmentally.鈥�

Oster agreed that it may be time to put aside reopening showdowns and instead work toward recovery.

鈥淭here is a very reasonable desire to move on from the discussion of, 鈥楬ow important were school closures?鈥� into, 鈥楬ow do we fix this?鈥欌�� she said. 鈥淚’m quite sympathetic to that desire to move on.鈥�

Eighth-grade math scores on the National Assessment of Educational Progress, often called the 鈥淣ation鈥檚 Report Card,鈥� fell by a jarring eight points since the test was last administered in 2019, while fourth-grade scores dropped by five points; both are the largest math declines ever recorded on the test. In reading, both fourth- and eighth-grade scores fell by three points, leaving them statistically unchanged since 1992, when NAEP was first rolled out. 

The findings comport with those of previous assessments of students鈥� COVID-era achievement, whether conducted by academic researchers or state and district authorities, which have shown undeniable evidence of diminished performance in English and especially math. Just a few months ago, the release of scores for 9-year-olds on NAEP鈥檚 鈥淟ong-Term Trends鈥� assessment 鈥� a different exam measuring today鈥檚 students against a baseline set in the early 1970s 鈥� offered similarly ominous results.

Even still, the education world has waited nervously for the unveiling of today鈥檚 data, perhaps the most important federal scores to appear since the pandemic began. Peggy Carr, commissioner of the National Center for Education Statistics, said that while relative stability in reading scores across some of the nation鈥檚 largest districts offered a few 鈥渂right spots 鈥� amidst all the chaos of the pandemic,鈥� the unprecedented reversals in math should spark serious concern.

鈥淣ormally for a NAEP assessment 鈥� we’re talking about significant differences of two or three points,鈥� Carr said on a Friday call with reporters. 鈥淪o an eight-point decline that we’re seeing in the math data is stark. It is troubling. It is significant.鈥�

A look at the results in their entirety show just how significant. There were no statistically significant gains in math, for either fourth or eighth graders, in any state in 2022. Instead, fourth-grade scores dropped significantly in 43 jurisdictions (either the 50 states, the District of Columbia, or schools operated by the Department of Defense Education Activity) while remaining statistically unchanged in 10. Eighth-grade math fell in 51 jurisdictions while holding steady in just two, Utah and the DoDEA schools. The average eighth-grade score has not only fallen since 2019 鈥� it is significantly lower than when the test was administered in 2003.

Translated into the exam鈥檚 performance levels, a massive downward shift can be seen. In 2019, 34% of fourth graders and 27% of eighth graders scored below the 鈥淣AEP Basic鈥� level in reading 鈥� the most rudimentary threshold of English mastery classified by the test. In 2022, those groups had grown to 37% of fourth graders and 30% of eighth graders, respectively. The below-basic classification also swelled in math, from 19% of fourth graders and 31% of eighth graders in 2019 to 25% of fourth graders and 38% of eighth graders in 2022.

Beneath the headline numbers, differing effects among student groups also made an impact on longstanding achievement gaps. For example, gaps expanded in fourth-grade math performance between white and African-American students, white and Hispanic students, male and female students, and students with and without disabilities. Conversely, gaps actually closed between many of the same groups in eighth-grade reading 鈥� including by a surprising seven points between English learners and native English speakers. 

Brown University economist Emily Oster, who has studied the effects of COVID and remote learning on student achievement, said that trends in NAEP scores were dynamic and varied, making them difficult to distill. Big-picture phenomena, however, broadly lined up with the existing evidence, she argued.

“Every state has four numbers, so one can construct quite a lot of different narratives around that. But the general patterns are that the losses are big, they’re much bigger in math than in reading, and they’re much bigger in more vulnerable kids. Those seem like things that are very consistent with every other piece of information that we’ve seen in post-pandemic testing.”

Julia Rafal-Baer is a K-12 education expert who serves on the National Assessment Governing Board, a nonpartisan body that sets policy for NAEP. In a statement, she said the results demonstrated the existence of 鈥渁n education crisis鈥� that demanded new solutions.

鈥淭he latest data isn鈥檛 telling us anything we didn鈥檛 already know,鈥� Rafal-Baer wrote in an email. 鈥淐OVID was exceptionally disruptive, and we鈥檙e running out of time to ensure that kids can indeed recover from this level of unfinished learning.鈥�

State-by-state comparisons difficult

No state could be said to have defied the downward pressure exerted by the pandemic and its countless challenges to learning. But the national averages do conceal substantial variation across different areas of the country. 

Some of the states where scores dropped the furthest, for example, were clustered in the mid-Atlantic region. Delaware鈥檚 fourth-grade math scores dropped an astonishing 14 points 鈥� nearly three times the national average 鈥� while its losses in fourth-grade reading (-9), eighth-grade math (-12), and eighth-grade reading (-7) were also significant. Virginia (-11 points in fourth-grade math), Maryland (-11 in eighth-grade math), and the District of Columbia (-8 in fourth-grade reading) also saw some of the worst declines across various grade/subject combinations.

By contrast, a small group of states seemed to weather COVID reasonably well, experiencing less severe declines than most. Overall, while performance in eighth-grade math was weakened virtually everywhere, 10 jurisdictions, including Georgia and Wisconsin, saw no statistically significant decline in fourth-grade math. Another 22 were able to stave off declines in fourth-grade reading, while 18 did so in eighth-grade reading. 

A small number of states 鈥� Alabama, Hawaii, Illinois, Iowa and Louisiana 鈥� kept scores from significantly falling in three out of four grade/subject combinations. Most impressive of all, Department of Defense Education Activity schools 鈥� 160 across 11 foreign countries, seven states, and two territories, each serving military families 鈥� saw no statistically significant drops in any subject or age group. Eighth graders in DoDEA schools, in fact, made the only statistically significant growth of any student group in this round of NAEP, improving in reading by two points since 2019. 

The differences between states will naturally raise questions about the procedures they followed to offer schooling during the pandemic. Among the states that saw the largest score declines, many stuck with remote learning far into the 2020-21 school year as a precaution against COVID spread. 

Oster, whose previous research has found that longer periods of remote instruction were linked with more severe learning loss, called the results 鈥渧ery consistent with what we’ve seen in state-level data, which suggests that places that had the most in-person learning lost less than the places that had more virtual learning.鈥� Even so, she added, a state like California 鈥� where she would have expected student scores to fall especially dramatically based on that correlation 鈥� instead saw more modest declines.

NCES鈥檚 Carr argued that the release provided little scope for comparisons between states, since so many jurisdictions experienced 鈥渕assive, comprehensive declines.鈥�

鈥淭here’s nothing in this data that says we can draw a straight line between the time spent in remote learning, in and of itself, and student achievement,鈥� she said. 鈥淟et’s not forget that remote learning looked very different across the United States 鈥� the quality, all the factors that were associated with implementing remote learning. It is extremely complex.鈥� 

Megan Kuhfeld, a researcher at the nonprofit testing organization NWEA, said that the average NAEP effects dovetailed with her own expectations based on of post-pandemic learning loss. That said, she agreed with Carr that the huge diversity of COVID learning policies 鈥� where neighboring school districts sometimes took radically different approaches 鈥� made direct comparisons difficult.

鈥� has supported the idea that remote instruction was a primary driver of widening achievement gaps, but I think it is harder to make that sort of inference at the state level because district reopening policies often varied widely within states,鈥� Kuhfeld wrote in an email.

Urban districts fared better in reading 

If a silver lining exists within the release, it comes from some of America’s biggest cities.

In addition to all 50 states and Washington, D.C., 26 urban school districts around the country participate in NAEP’s Trial Urban District Assessment program. The measure offers a unique look inside districts that collectively enroll millions of students and were subject to substantially different state-level public health policies.

Disappointingly, math results in these districts were no better than elsewhere: Fourth-grade and eighth-grade scores alike sank by eight points on average, matching or surpassing the declines for the nation as a whole. 

Performance in English, however, offered somewhat sunnier news: Average scores in reading held up in 17 cities, falling in just nine. Fully 21 of the 26 urban districts managed the same in eighth-grade reading, with only Shelby County, Tennessee; Jefferson County, Kentucky; Guilford County, North Carolina; and Cleveland, Ohio, experiencing statistically significant drops. In Los Angeles, the nation鈥檚 second-largest district, eighth-grade reading performance even improved. 

Michael Petrilli, who leads reform-oriented Thomas B. Fordham Institute, nevertheless took a dark view of the overall NAEP outcomes. 

“There’s no sugar coating these awful results,” Petrilli said. “Save for Los Angeles (which I honestly cannot explain), the only question is whether states and localities did bad or worse. These data tell us how big a hole we’ve dug for ourselves. Now it’s up to all of us to dig ourselves 鈥� and our students 鈥� out.” 

Others took a somewhat more hopeful outlook. Tom Loveless, a longtime education researcher who formerly led the Brookings Institution’s Brown Center on Education Policy, said the urban districts鈥� results provided “a glimmer of hope in these otherwise dismal data.” Moreover, he added, both the NAEP data and state standardized test scores have already shown evidence that student achievement is bouncing back from its pandemic nadir.

Going forward, Loveless observed, state and school district leaders will likely view this round of scores as a kind of new student performance baseline. That could provide an accountability mechanism if things don’t improve.

“I think 2021 was probably the bottom, and we’re getting little shards of progress in these NAEP data,” he said. “But I’m expecting [the 2024 NAEP results] to look quite a bit better, and the state tests, too. If they don’t, I think people will start raising harsh questions.”

But the results from Curriculum Associates, which publishes the I-Ready assessments, also reveal how much work remains to be done: Fewer children in the early grades are developing essential phonics skills, they found. In fact, even more were below grade level in the spring of 2022 than there were a year ago 鈥� in third grade, 27%, compared to 24%.

The results in math reflect a similar trend. Performance for some students is not only below what it was in the years before schools shut down; it鈥檚 worse after a year of mostly in-person learning. In grades five through eight, for example, fewer students than ever are developing essential math skills like understanding place value, multiplication and fractions. 

鈥淭hat first year back, they didn’t recover all the way,鈥� said Kristen Huff, vice president of assessment and research at Curriculum Associates. 

Last year, teachers had more students below grade level than ever, Huff said, and the 鈥渨hole classroom dynamic changed.鈥� Foundational math and reading skills 鈥渁re the two bridges where we’re seeing those gaps endure.鈥� 

The findings follow the recent release of long-term achievement data from the National Assessment of Educational Progress, which revealed sharp declines in reading and math for the nation鈥檚 9-year-olds. I-Ready, with results in first through eighth grade, serves a different purpose than the so-called 鈥渘ation鈥檚 report card.鈥� But the takeaways are similar: The pace of recovery is slow, and those who were struggling before the pandemic have the steepest hill to climb, particularly Black and low-income students.

Among I-Ready鈥檚 more encouraging signs, however, is that schools serving a majority of Black and Latino students saw the greatest gains in fourth grade in both reading and math. From a sample of almost 5,000 schools, over 300 exceeded expectations, despite serving students at least two years behind. Experts say the results show the need for well-targeted academic support.

鈥淲e don鈥檛 necessarily say everyone needs acceleration in all subjects for the next five years, but some kids will,鈥� said Libby Pier, chief of staff at Education Analytics, a Madison, Wisconsin, nonprofit that monitors state data. 鈥淭here are certain skills, certain students and certain schools where things are looking less or more dire.鈥�

鈥榃anted the data to tell us鈥� 

In the testing field, I-Ready is known as a diagnostic assessment 鈥� used to identify what students know and where skill gaps remain. After states canceled annual tests in 2020, schools relied more on I-Ready and similar assessments to better understand the pandemic鈥檚 impact. 

鈥淭hey have continued to play an important role in providing more timely information to schools about recovery efforts,鈥� Pier said. 

At Blakesburg Elementary School, among those s where students scored better than expected, Principal Tammy Davis began using I-Ready to monitor students鈥� reading comprehension skills and supplement a state test that focuses on fluency.

The school is located in a farming community in southeast Iowa. About 30% of students are low-income, and even though the school distributed Wi-Fi hotspots, many students lacked reliable internet during remote learning. After seeing 鈥渉uge dips鈥� in reading and math when schools reopened in the fall of 2020, teachers examined every student鈥檚 score on each category of the assessment. 

鈥淚 had a feeling where we were the lowest, but I wanted the data to tell us,鈥� Davis said.

To build comprehension skills, teachers never stopped devoting time in the school day to social studies and science 鈥� even though many districts prioritized math and reading, said April Glosser, curriculum director in the Eddysville-Blakesburg-Fremont Community School District.

鈥淚f you鈥檙e reading something that you have nothing to connect to, it鈥檚 never going to make sense to you,鈥� she said, pointing to a passage on the sport cricket. 

Teachers, she added, began to do more 鈥渢hink alouds鈥� explaining  to students what to do if they ran into words or phrases they didn鈥檛 recognize or understand. The school鈥檚 sixth graders, who have always struggled, Davis said, saw significant growth, from 35% proficient in spring 2021 to  56% proficient last school year. Schoolwide, 74% were proficient, getting closer to the goal of 80%.

Emerging state results

Students at Blakesburg spent most of the 2020-21 school year in person, but that wasn鈥檛 the case in many other districts. Asstates begin to release test results from last school year, researchers plan to further examine links between remote learning and student achievement.

Also Tuesday, Brown University economist Emily Oster released of state test data confirming that learning loss was greater in districts that had more remote and hybrid instruction.

In Virginia, for example, just 5% of students attended districts with 鈥渉igh levels of in-person instruction鈥� during the 2020-21 school year. In 2021, the percentage of students reaching proficiency in math and English language arts was 20.5 percentage points lower than in 2019. This year, it was still 12 percentage points lower.

In Louisiana, almost two-thirds of students learned mostly in person in 2020-21. Proficiency levels in 2021 and 2022 were 5.5 and 4.1 percentage points below 2019, respectively.

鈥淒istricts that had more remote learning during the pandemic have a much longer way to go,鈥� Oster said in a statement.

Kamenetz had spent years covering the heaviest stories on the education beat as an award-winning reporter at NPR, from the destruction wreaked by Hurricane Katrina on New Orleans classrooms to the potential risks of excessive screen time. And according to her sources, the coming pandemic wasn鈥檛 just going to drive down math scores or disrupt the teaching profession: Prolonged school closures would leave a mark on child safety, mental health and social development. 

The months that followed, shaped by academic stagnation and political division, frame Kamenetz鈥檚 new book, The Stolen Year. Released Tuesday by PublicAffairs, the volume reads like a reckoning with the predictions made by the experts she consulted more than two years ago. Each chapter examines a facet of social policy in America that was fundamentally challenged by the emergence of COVID-19, from the courts system to K-12 schools, and the effects that were felt by tens of millions of children.

And how much of her sources鈥� collective warning was validated?

鈥淎ll of it,鈥� she told 蜜桃影视.

Much of the social toll, measured in deaths or distance or deterioration of services, was unavoidable. But Kamenetz argues that the failure of online learning, and of in-person schools to reopen faster in thousands of districts, was also highly contingent 鈥� on leaders鈥� failure to adjust during the fateful summer of 2020, but also on experts and members of the media, whose message to the public was too often muddled. Though every Western country had to scramble to come to grips with a once-in-a-lifetime public health emergency, few kept children out of school longer than the United States. And virtually none were as divided in their political and policy responses.

Kamenetz observed those responses as a veteran journalist, but also as a mother of young children and a parent in New York City, where the official response to COVID was often scattershot. That experience 鈥渃omplicated鈥� her view of public schooling in this country, she confided.

鈥淚 very much understand the perspective of people who feel betrayed by public schools, wherever they’re coming from politically,鈥� she said. 鈥淭here’s been such a fraying of the consensus around what is really our major piece of social infrastructure for families.鈥�

This interview has been edited for length and clarity.

蜜桃影视: What’s it been like reporting on the pandemic for over two years, including writing this book, as both a veteran education reporter and a mother of young kids?

Anya Kamenetz: Over that weekend after March 13, the day that schools started to close, I was trying to confirm a coronavirus case in my child’s school, which I was then going to report to WNYC [New York City鈥檚 public radio station] and NPR. It’s not regular for reporters to report on their own kids’ schools, but this was obviously of import.

For a couple of weeks, I’d been covering the college shutdowns in Asia and then in the U.S., which were a precursor to the K-12 shutdowns here. But when de Blasio made the call to shut down schools 鈥� which he did, with his characteristic leadership, very late on a Sunday night after saying he wouldn’t do it 鈥� that was really the beginning of it for me. And I knew it was going to be a very big deal because I had not only been at NPR for six years at that point, but I’d covered [Hurricane] Katrina, which I believed would be a decent parallel: You have this society-wide catastrophe, and within that, kids are pulled out of school amid all this instability and trauma. The impacts of that were devastating, and the empirically measured effects on young people in New Orleans were still there 10 years later. 

So my job was to document what was happening, and it was a very special position to be in with my skills and prior research. I also had an interest in refugee education and what is called “education in emergencies.” So my contacts in international development circles were some of the first people I called up to ask, “How’s this gonna go? What should I be looking out for?” And everything they told me panned out.

You’re originally from New Orleans, right?

Yes.

Your instinctive Katrina comparisons have unfortunately proven accurate 鈥� test score data from NWEA shows that learning loss during the pandemic is pretty comparable to the damage suffered by students in New Orleans after the hurricane.

From what I can tell, it’s actually worse. The latest NWEA data indicates that the average elementary student is on track to recover in three years; for Katrina, it was two years. And after Katrina, those kids went back to 鈥渂etter schools鈥� 鈥� better funded schools, certainly, and schools that achieve higher test scores.

That’s not what we’re seeing here. In the July 2022 NWEA study, it found that middle schoolers weren’t making any gains. We don’t even have a trajectory for them. 

The predictions in the research pointed me to the conclusion that high school students were the ones to worry about because they were on a course to separate from school. It’s not a question of catching up, it’s a question of whether they were going to stay in school. That is something we should worry about, especially with college-going .

Getting back to the early guidance offered by your international sources: How many of their predictions panned out?

All of it panned out.

The idea that young people would go into paid work and that young girls would become caregivers; that the impact of school closures would match underlying inequalities in society, and that children who were privileged enough would see no difference; that continuity efforts to keep learning on-track 鈥� this stuff goes back to World War II, for instance, when the BBC had its own children’s service that carried school broadcasts 鈥� don’t reach everyone, and in fact multiply inequality because the kids who can benefit from those efforts are already advantaged. All of that was exactly what we saw.

Those insights were remarkably prescient. In my own reporting, there were also some dogs that didn’t bark, most notably some early coverage on the possibility of a Great Recession-type crunch on school finances. 

That’s right. There was a short-term spike in child hunger 鈥� it could have been two or three months 鈥� and we have to remember that even one month of increased hunger is very bad. But kids actually finished the end of the year with money in their pockets, so the economic picture is a little complicated. And there wasn’t the kind of housing crisis that left huge percentages of kids homeless, which is something that can happen after natural disasters. 

In fact, if you get into , you can see that between unemployment and CARES Act money, families spent more time at home with their kids. There’s obviously a segment of society that doesn’t have paid leave. Patricia [one of several parents profiled in The Stolen Year], who’s a mom in D.C. with two kids, worked in D.C. Public Schools; she had her child prematurely in July and then went back to work in August. So during the pandemic, it was the only time she was able to be home with her children. As hard as it was, she saw the advantage in it.

What comes through in the book is that there was a lot of contingency, particularly in those first pandemic months. Do you think there was a way that the initial COVID wave could have been handled better by education authorities, either at the local, state, or federal levels. Could things have gone significantly better than they did?

School closures were sort of handled in the spirit of President Trump when he said, “We’ve got to shut this whole thing down until we figure out what the hell’s going on.” They were not controversial in that early wave because it was happening all over the world, and people really needed a handle on what was happening. The first big mistake was not making a plan, as soon as we shut them down, of how we were going to open them back up. 

We had this false idea about a two-week turnaround, and people really froze and didn’t plan for the future. If you’re on this beat, you know that schools start planning for staffing and scheduling in the next academic year by April. And that was when schools in parts of Europe started reopening, in April and May. They were like, “This is truly an emergency measure, and we’re going to treat it as such. We’re not going to let it continue through inertia because if districts don’t have clear leadership and messaging, they’re not going to respond.” And that’s what we saw here.

It’s unclear to me what the academic effects were of not reopening schools to finish out the 2019-2020 school year, as opposed to the extended virtual learning that took place the following year. I suspect the bigger impact was the precedent of closing through the end of the year, which seemed to set us on a course toward virtual learning being the accepted way to deal with this.

I agree completely. The effect of reopening in May 2020 would have been to put the training wheels on and get everyone onboard with the fact that this was going to happen. It was described to me by a teacher in Florida who went back into the classroom in 2020. She was like, “I know colleagues who stayed home with medical exemptions, and when they came back, they were terrified. So I said, ‘Listen, it’s not that big a deal, we’re doing this.” 

What was absent was the comfort of everyday routine, and the sense of control that teachers can have when they have protocols. By not opening in the spring of 2020, we made it that much harder. And there was a slice of schools in blue states that didn’t open up until February, March, April 2021, and it got harder and harder the longer they waited.

It seems like what set the U.S. apart was the diversity of local responses we saw in the summer and fall of 2020. Of course, there wasn’t a ton of clear federal guidance, and God knows if blue states and districts would have taken that from the Trump administration, but it yielded this fractured approach to reopening and public health measures that carries on to this day.

Writing the book, I felt compelled to do this deep dive into the history of our public school system. Because it’s very different from our peer countries, which have a national system and a national curriculum and a minister of education who actually runs the schools. Instead of that, we have all these different districts.

All of which is to say that there was an absence of guidance from the top, but also an absence of data collection. I covered a study from Spain in September 2020, which tracked case rates by region. Every region had all the same protocols. After schools reopened, cases went up in one region, down in another region, and stayed flat in some others. So they released this paper and said, “We don’t think what you’re doing in schools is making a big difference, please carry on.” The CDC didn’t release an equivalent paper until, like, January [2021]!

Where was the information? 鈥� this mom-and-pop data company that was started for totally different reasons 鈥� became the nation’s go-to source for what schools are doing because the federal government just wasn’t tracking and releasing enough information. 

You mentioned Bill de Blasio’s sort of tortured initial move to close schools in New York City, which attracted a lot of criticism. But from the distance of two-plus years, I suspect that many observers wish that districts had been more deliberate in shuttering schools and keeping them shuttered, even if that’s clearly a case of Monday morning quarterbacking.

That’s right. More forward-thinking, more innovative leadership could have gone a long way. And as frustrating as it often was to live in de Blasio’s city, it’s worth remembering that of the big blue cities, they opened schools up first. Though they had really bad uptake of hybrid [learning] because it was untenable from a child care perspective, and also because they delayed reopening twice. 

People don’t understand that there’s a dynamic between parent trust and how you communicate the decisions you’re making. The more you hem and haw and quibble over things, the harder it is for parents to actually trust that you’re going to do the things you say you’re going to do. One lesson from this period is that being a bad communicator is a problem in and of itself.

The consequences of that miscommunication and opacity in decision-making are already being felt politically, with school board members enduring a wave of recall attempts and Democrats losing ground on the issue of K-12 education. I’m wondering if your own levels of trust in political leaders and education authorities have diminished.

The process of reporting this book has complicated my view of public schooling. 

That’s a pretty broad thing to say, but basically: I thought of public schooling as a public good and something that needed to be freely available to all people, and I still think that. But I also very much understand the perspective of people who feel betrayed by public schools, wherever they’re coming from politically 鈥� whether they think public schools are racist, or that they’re not representing their family’s point of view, or whatever. There’s been such a fraying of the consensus around what is really our major piece of social infrastructure for families. In terms of the , a lot of parents feel like they were forced to exit. 

Of course, time will tell. There’s a powerful pull back to normal, and some families are changing what they’re doing. You see a lot of expressions of relief about coming back [to school]. But from what I understand of , both politically red districts and districts that reopened sooner 鈥� those are often one and the same 鈥� are reporting a rebound in enrollment. Politically blue districts, and those that opened up later or kept mask mandates in place, are continuing to lose students, which is what we’re seeing in New York City. That’s more important than what people say in polls, I think.

I take your point that school closures were mostly uncontroversial when they were first enacted. From my recollection, the point where some of this dissatisfaction began to set in was in fall 2020. According to Burbio data you cite in the book, 42 percent of students returned to all-remote schools that September. 

I actually think it’s more. The federal government released a report in February 2021 breaking things down schools by what they offered: in-person, fully remote, or hybrid. But we know that within the in-person and hybrid categories, there were also families that chose remote instruction. In that study, around half of all students were still fully remote that February, so it’s pretty safe to say that a majority of kids were in remote-only schools in fall of 2020; probably only about one-third had an in-person option.

In New York City, to my knowledge, about one-third of kids were actually showing up to school at that time. And that was an 鈥渙pen district.鈥�

Okay, that’s an important caveat. Whatever the exact figure was, massive numbers of kids didn’t have the option of attending school in-person, even months after some comparable countries had fully switched back. Why?

They had not planned to come back in-person. There was a statement by Gov. Gavin Newsom in summer of 2020 about potentially opening up for summer schools; as we know, California schools didn’t reopen until spring 2021. So they just didn’t do the planning necessary. We might also say that states didn’t have the necessary resources in terms of public health tracking and contact tracing. I spoke to a woman employed as a contract tracer in New Jersey, and her experience in that job was why she didn’t send her kid to school. She was like, “This isn’t working.”

The testing also wasn’t in place in 2020, and obviously, there was opposition from people who didn’t believe their workplace was going to be safe. Some of the opposition came in the form of bringing cardboard coffins to marches and saying that children were going to die. That was not supported by the evidence, but it was scary.

The messaging was very heated on both sides of the reopening debate. What I don’t really remember was a more dispassionate accounting that weighed the legitimate public health concerns of both families and school employees against the legitimate educational and social needs of kids. 

Absolutely. That was not done. And we can see that from the fact that there was no real attempt to triage the situation. 

San Francisco tried to do this. They said, “We have lists of kids who are in foster care, kids who are in substandard housing, kids who are recent immigrants, and kids who are disabled, and we’re going to prioritize them for inclusion in learning hubs.” Okay, great. But they only created half of the hub spaces they said they would. The failure is in not balancing what you’ve articulated as important needs, and instead allowing chaos in the market to take place instead of those things.

You mentioned child care, which was a troubled industry even before the pandemic. In the book, you memorably refer to a “laser maze” of obstacles to access care. What did the pandemic teach us about how the United States provides services in this area?

The pandemic made it really obvious that we have no infrastructure for care in this country, and in fact, it would be exaggerating to call it a system. There’s been research showing that a certain percentage of child care providers use their own food stamps to feed kids in their care because our subsidy system doesn’t work the way it’s supposed to. 

We also learned that it’s essential infrastructure, and people can’t go to work without child care. What we rely on for young kids is a gradient of unpaid and underpaid care. COVID made a certain amount of people recognize that, and I was very happy to see some of these ideas [for child care and pre-K subsidies] in Congress. But seeing them yanked away in this version of is really disheartening and makes you ask what it’s going to take for politicians to change this. 

Among the big priorities the Biden administration wanted to address in January 2021 鈥� and even including some that weren’t big priorities then, like inflation 鈥� most have been acted upon. Not child care or pre-K.

Yeah. Political theorists would probably say that the climate movement pushed their issue to the top of the agenda. Health care for seniors has always been a win for politicians, and there’s a clear constituency of people who vote a lot on that issue. But when do moms have time to march? When can they crowd into Nancy Pelosi’s office and make it unavoidable for this to change? I’d love dads to do this too, but my point is that it takes organizing.

We both cover education research. How do you grade the performance of education experts and policy specialists, many of whom gained big microphones over the last few years? You brought up Burbio, which I definitely hadn’t heard of before 2020, but I was thinking of figures like Brown University economist Emily Oster: people who were not affiliated with government but took it upon themselves to gather what information was available and communicate it to the public.

That question is going to be the subject of a lot of research and reflection going forward: How does this ability to speak directly to the public affect how research is conducted, how it’s publicized and how it shapes public opinion?

We talked about the vacuum of guidance from the federal government in the early stages of COVID, which kind of distorted the whole information ecosystem. And then the virus kept changing in really uncertain ways such that Omicron acted differently than Delta, which acted differently from the first wave. People really wanted certainty, and there wasn’t certainty to be found, so it was a foregone conclusion that people peddling reassurance were going to get a lot of attention. 

Do you have anyone specific in mind?

My previous book was about screen time, which is a hotly contested area without as much data as you’d like. It’s also a situation that’s always changing because new forms of media emerge. Something I found very useful during the pandemic was just being able to fairly convey that truth to parents that just want information to navigate their day. You have to be very clear about what you don’t know and how to make a decision depending on what your specific concerns are.

I think that was why someone like Emily Oster got so popular. Her wasn’t telling people what to do, it was a framework for making decisions.

For me, the urgent need for information made the pandemic a unique period of connection between reporters and readers. I imagine you were getting a ton of feedback, both positive and negative, from parents on Twitter as the various COVID debates raged on.

The uncertainty was so excruciating that it led people to retreat to their corners and take refuge in what their tribe was doing. That’s why there was this subset of people who were like, “I’m a liberal mom who thinks my toddler shouldn’t have to wear a mask anymore. But people hate me for that and think that I’m a Trump supporter.” 

Because it was so polarized, it was very hard to deal with these gaps. Or sometimes there weren’t even gaps! It was more like, “You have a child with a speech impediment, but we have a grandparent at home, so our concerns are different.” There was a need to give people a little more grace.

I’d add that we’re education reporters, not science reporters, and when we had to call up epidemiologists to get your story, there was a lot of caution around that. It was hard for a lot of reporters, and that led to gaps in coverage: We were comfortable talking about what was happening academically but not as comfortable talking about public health. For that, we really listened to the health experts, and the priority of the health experts was preventing even one case of COVID. With a little more confidence, we could have taken a broader view and said, “We might be reducing COVID spread by x amount, but what’s happening to these kids at home?”

This wasn’t even a phenomenon restricted to the education press. In spring 2020, you could get the feeling that the only relevant expertise was in health. It almost would have been strange if reporters didn’t become deferential around experts in that field. 

“Deferential” is the exact right word. And the solution for it was for education reporters to stay in our lane and report on what was happening to kids. I realized early on that NPR didn’t have a child care beat; we didn’t have a child welfare reporter, so we didn’t really know what was happening as far as kids getting abused and not being flagged by a mandatory reporter. 

We just had a schools desk with three reporters for the whole country, and one was doing higher ed. That wasn’t enough to cover what was happening to kids, particularly when schools were closed.

It takes so long to pitch, write, and edit a book like this. I’m curious how much changed after you submitted your first draft to the publisher, and if you had to rewrite a lot of this as conditions on the ground evolved.

I would say that trend lines basically continued, but the closures of the Omicron winter really impacted learning in 2021-22 鈥� so much so that I don’t think it’s really fair to call it a recovery year. The amount of closures and quarantining and chronic absenteeism were too grave to say there was a huge amount of recovery, as opposed to just creeping back to normal. 

The school leaders I know are just beginning to contemplate what a full, normal year might look like, and is being framed accordingly. They’re very clear now 鈥� now 鈥� in saying that kids should be in school at all costs. In that way, Ifeel like public opinion and people’s experiences have evolved to the point that they’re ready to have this conversation. And I hope we do have it; I hope we don’t rush to saying, “Why aren’t you over that yet?” or, “There’s a huge achievement gap, why haven’t these kids caught up yet?” Like, let’s not forget what happened here.

So is it possible for 2022-23 to be the first post-COVID year? And what’s it going to take to make that happen?

I do think it’s possible. But in order for that to happen, we need to be clear-eyed about what has already happened. It’s a little frustrating to see schools lurching from crisis to crisis, and there’s a crisis rhetoric around schools that doesn’t always match reality. We’re hearing about a teacher shortage, for instance, but there were teacher shortages before the pandemic. And also, schools are listing a lot more vacancies; it’s not so much a shortage as schools trying to hire more people.

So there is a chance for recovery. In order for there to be recovery, schools need to do what they say they’re going to do in their [American Rescue Plan] plans and not lose sight of their responsibility to help the most vulnerable and the kids who lost out the most in this pandemic.

The premise of your book was that a year of learning was stolen; what needs to be done to restore what was lost?

You have to hear, without interrupting, what harm was suffered. You hear what kids went through. And then you try to give them back what they lost. That’s going to take time, but it can be done. A wonderful thing about children is that they have time, and the investment you make today will pay off many times in the future. We just need to give them that chance.

See previous 74 Interviews: Harvard economist Thomas Kane on reversing COVID learning loss, Seattle superintendent Susan Enfield on 700 days of learning disruptions, and Arizona assistant principal Beth Lehr on the pandemic鈥檚 effect on teachers. The full archive is here. 

Along with collaborators from Dartmouth, the CALDER Center at the American Institutes for Research, and the nonprofit testing group NWEA, Kane released incorporating pre- and post-pandemic testing data from over 2 million students in 49 states. Its conclusion: Remote instruction was a 鈥減rimary driver of widening achievement gaps鈥� over the last several years, with schools serving poor and non-white students suffering some of the greatest setbacks.聽

Overall, Kane and his co-authors found, high-poverty schools were more likely than others in the same district to stay remote throughout the 2020-21 school year; among all schools that stayed remote for longer, students at high-poverty schools showed much worse declines in math scores. And they calculated that some school districts would have to spend every dollar of their federal COVID relief money on academic recovery efforts to have any hope of making up the lost ground.

As Kane observed for the Atlantic, local education authorities are required to use only 20 percent of those funds on pandemic-specific remediation. And there is sufficient reason to doubt that even the most promising educational interventions, such as personalized tutoring, can be delivered at the necessary scale to reverse the damage inflicted by COVID. Even the Biden administration鈥檚 recently announced campaign to recruit 250,000 new tutors and mentors is at least several months away from being fully realized.

Kane, the faculty director of Harvard鈥檚 Center for Education Policy Research, has spent decades carefully evaluating the effectiveness of school improvement efforts. A Council of Economic Advisors staffer during the Clinton presidency, he has studied school accountability systems, teacher recruitment policies, and the effects of affirmative action throughout long stints in both academia and think tanks like the Brookings Institution. His research on teacher evaluation inspired a half-billion-dollar initiative launched by the Bill & Melinda Gates Foundation to lift classroom performance in large school districts around the country.

Now he鈥檚 hoping to work with state and district leaders to combat an educational disaster whose effects, he says, are still not well understood. While policymakers may now have a loose idea of the challenges facing educators and families, the policies they鈥檙e currently reaching for will likely prove inadequate as a solution.

鈥淥nce that sinks in, I think people will realize that more aggressive action is necessary,鈥� Kane said. 鈥淚n the absence of that, it’s hard to blame local folks for not taking more aggressive action because they have no way to know that what they’re planning is nowhere near enough.鈥�

This interview has been edited for length and clarity.

蜜桃影视: How do your findings and research design differ from earlier studies that have looked at pandemic-related learning loss? I鈥檓 thinking specifically of last year鈥檚 study conducted by, among others, Brown University鈥檚 Emily Oster, which also pointed to really steep losses associated with the switch to virtual learning.

Thomas Kane: There are at least two ways that this paper is different. The first is that we’re able to estimate magnitudes of losses in a way that’s comparable to the effect sizes of [educational] interventions. In that [Oster] study, they can focus on changes in proficiency rates on state tests. Each state has its own cut score, so the magnitude of the changes in proficiency rates depends on whether that cut score is near the middle or near the tail of the test score distribution. If my cut score is near the middle, even a small decline in achievement can mean a big swing in proficiency. But if my cut score is at the tail, even a large decline in test scores can show up as a small change in the percentage of people who are proficient.聽

So while that study could qualitatively describe what was happening 鈥� in areas that were remote for longer periods, proficiency rates declined 鈥� they really couldn’t characterize the magnitude of the decline in any way that was comparable to the effects sizes, which I think is critical. As we’ve argued, it’s not at all surprising that there were larger losses in places that were remote for longer periods. It’s the magnitude of the losses that’s startling.

This design also lets you make comparisons within districts, as well as between districts, right?

That’s another big difference between our paper and what’s out there now. The [Oster] paper was focused on district proficiency rates, and what they found was that districts with larger shares of minority students and high-poverty schools had larger losses. But it could have been, for instance, that the implementation of remote learning was just weaker in those districts 鈥� districts with a higher share of students in poverty may have seen bigger declines in achievement, but the losses could have been similar in the high- and low-poverty schools in those districts.

By being able to look within districts, we were able to test whether the number of weeks of remote instruction had disproportionate impacts on high-poverty schools and minority students in those districts. Our answer was pretty clearly yes, there were bigger losses. And it wasn’t just because the urban districts had a harder time implementing remote instruction; even within those districts, the higher-poverty schools lost more.

You used the word “startling” to describe the learning loss. Were you expecting to see effects of this size?

We went in without any clear expectations on the magnitude of the impacts we would see. The reason why I called it startling was because I know that there are very few educational interventions that have ever been shown to generate a .45 standard deviation impact [a common measure showing the difference in any population from the statistical mean; they can be loosely converted into other units, such as learning time or dollars spent] on achievement. Yet that’s the size of the loss that high-poverty schools that were remote for more than half the year sustained. So it was startling because when we compare the impact estimates of remote learning to the potential impact of the available interventions, it’s clear that there is no one thing that we could say, “If all districts did this and implemented it with fidelity, it would eliminate the gap in one year.”聽

For instance: In a review of the pre-pandemic research, tutoring has been found to generate a gain of about .38 standard deviations. Well, you could provide a tutor to every single student in a high-poverty school that was remote for half the year and still not close the gap. You could get close, but you wouldn’t close that gap. And we know that districts are never going to be able to hire enough tutors to provide one to every student in a high-poverty school, let alone deliver that tutoring at the level of quality as these programs evaluated in the research. That’s why it was startling 鈥� not just because it conflicted with our prior expectations, but because when we saw it, we realized that we couldn’t come up with a long list of interventions that yield effects of this size.

So what can schools and districts realistically be expected to do in this situation? 

We can’t be thinking of this as a one-year catch-up. If we really are committed to making students whole and eliminating these losses, it’s going to be multiple years. There are other interventions that have been shown to have effects, it’s just that no single intervention gets you all the way. 

One example is double-dose math. There’s , and , that found that an extra period of math instruction over a whole year generates about .2 standard deviations. 

So more districts should probably be thinking about something like that, especially in high-poverty schools. But like tutoring, increased math instruction requires staff; you can’t double the number of math classes students take without increasing the number of math teachers. Again, districts should be considering doing some of that, but it will also have constraints on the scale they can implement. 

Another possibility, which a lot of districts are already planning for, is summer school. There are studies suggesting positive impacts of summer school. But [the effects are] small. The big challenge with summer school is getting kids to attend regularly, because it’s viewed as optional learning time. That’s not a reason not to scale up summer school, it’s just that we shouldn’t think that doubling or even tripling the percentage of kids going to summer school is going to close these gaps. It’s not. You get a learning gain of about .1 standard deviations 鈥� around five weeks of learning 鈥� based on the pre-pandemic research.

One option that really hasn’t gotten much serious consideration, largely because of political pushback from parents and teachers, is extending the school year. If we extended the school year by five weeks over the next two years, that would obviously cover 10 weeks of instruction. I recognize that teachers would have to be paid more for that time. In fact, they ought to be paid something like time-and-a-half. But that’s the kind of option that I hope will gain attention once people realize the inadequacy of the steps that they’re currently considering, like small increases in summer school or tutoring a small percentage of students. It’ll become apparent that that’s just not enough, though my fear is that it may not become apparent in time. Based on what I’m seeing, most districts are going to find that students are still lagging far behind when they take their state tests in May 2023. The danger is that if they only discover that then, and only start planning more ambitious recovery plans then, much of the federal money will have been spent already. That’s why we’re trying to get the message out about the scale of the declines, and the likely scope of the efforts required to close them, while there’s still time to act.聽

Districts only need to spend 20 percent of their COVID relief funds mitigating learning loss. But you and your co-authors created a formula to determine the financial cost of reversing this academic harm, and in many cases, that figure would basically demand every dollar allocated by Washington.

We try to put the scale of the [learning] losses and the amount of aid that districts have received in the same scale. We report both as a share of an annual school district budget, which I think is a useful starting point for thinking about what it’s going to cost a district to recover. If a district has lost the equivalent of, say, 22 weeks of instruction as a result of being remote, and you’re asking what it’s going to cost to make up for that, the lower bound of the estimated cost would have to start with [the question], “What does it cost to provide 22 weeks of instruction in a typical school year?”

The answer would be whatever share of a district鈥檚 typical annual budget is spent over 22 weeks. In the paper, we use a 40-week year, under the assumption that salaries are paid over 40 calendar weeks instead of just 36 instructional weeks. And then we put the amount of federal aid that districts got on that same scale 鈥� say, what share of a typical year’s budget districts receive. We think that’s a useful starting point for people, and what they’d see is that in the high-poverty districts that were remote for more than half of 2021, the amount of aid they received is basically equivalent to 鈥� maybe a little more, but not much more than 鈥� the magnitude of their losses in terms of instructional weeks. That just means that, rather than spending the 20 percent minimum that was required in the American Rescue Plan, some districts should be thinking that they’ll need all of that aid for academic catch-up.

I have to say, this conversation is leaving me pretty pessimistic that some of this lost ground can ever be fully recovered. Without asking you to look into a crystal ball, is that concerning you as well?

Yes, but here’s a more hopeful spin: A friend of mine sent me a political ad for one of the gubernatorial candidates in Rhode Island, Helena Foulkes. She says, “I’m running for governor, and my top priority is restoring students’ achievement, and if I fail to restore achievement, I’m not going to run for reelection. Hold me accountable for whether we catch kids up.”聽

I would hope more politicians take that pledge, and that the way to judge mayors and school board members and governors over the next couple of years is on whether they succeed in restoring students to their pre-pandemic levels of achievement. It would be that kind of accountability that would wake people up to the need for more aggressive action now. It’s one thing to read these reports about achievement losses nationally, but it’s another thing to see that your own schools, locally, followed exactly the pattern of this report. 

Most districts have seen the statistics from [the Oster paper] and know that their proficiency rates have declined by 10 or 15 percentage points. But that kind of statistic, as we’ve discussed, doesn’t really convey the severity. We’d like to provide districts with the tools to gauge the losses in the kinds of units 鈥� like standard deviations, or dollars spent, or weeks of instruction 鈥� that they could compare to the effect sizes of educational policies. That could make it easier for people to translate their local losses into a package of interventions of equivalent size. In , I tried to put both the learning loss and the intervention effects into instructional weeks rather than standard deviation units to make it easier.

I think that needs to happen. Local decision makers need to see the scale of their students’ losses in ways that are more readily comparable to the expected effect sizes of the interventions they have to choose from. Once that sinks in, I think people will realize that more aggressive action is necessary. In the absence of that, it’s hard to blame local folks for not taking more aggressive action because they have no way to know that what they’re planning is nowhere near enough. It certainly sounds impressive to say, “We’re going to double our summer school enrollment and provide a tutor to 5 percent of the students in our schools.” All of that would reflect more than the catch-up effort in a typical school year, but it’s only when you compare those to the effect sizes for those interventions, and the magnitude of their losses, that you realize that it’s nowhere near enough. So we’ve got to make that lack of proportionality clearer to local decision makers, and not just in these national reports.

Another recent study using MAP data found that U.S. students had sustained as much academic damage from school closures as kids in Louisiana suffered after Hurricane Katrina. But after the storm, the whole New Orleans school district was fundamentally restructured, such that it’s now mostly composed of charters. What do you think of more drastic attempts to change the organization of schools and districts? 

Here’s one reason why this challenge is greater 鈥� and it’s actually related to the situation in Boston. I think that if people were confident that a state takeover would produce the big improvements that are necessary in Boston, there would be political will. The problem is the uncertainty: “If we take this very difficult step, is it going to produce the results we’re hoping for?” 

If some district said, “We’re a high-poverty district, and we were remote for more than half of 2021. What should we do?” I could list a few things they should be trying, but I couldn’t point to a package that would definitely close the gap because it’s an unprecedented gap. There is one thing I think could provide the hope and ammunition that would generate political will: We could organize for the next few months around a set of interventions to be launched in the spring of 2023 and then find a few places that would be willing to try that package of things. If we could evaluate those and generate some results early in the summer of 2023, we could then say, “Here is a set of interventions that, if you implement them, it’ll get you a long way toward closing the gap.” And I think we’d have an easier time persuading people to use the political capital you need to invest in that.

So to anyone reading this interview: If there are districts or states that are willing to implement some really creative catch-up strategies next spring and want to contribute to an evidence base that the rest of the country can use, I want to work with you! Right now, there’s no package of efforts that I’d be confident would be enough to close the gap. Absent that, it’s no wonder that politicians aren’t willing to invest political capital in it. But if we had that, we could all get behind advocating for them. It would help everybody if a small set of districts would step forward and try to provide a model for the rest of the country to copy.聽

The clock is ticking, and I think we’d have to do it next spring. I’m sure we could design a study and get results out quickly to people about the type of effort that would generate enough [learning] gains. But there shouldn’t be just one model we’re trying 鈥� there should be multiple approaches that we systematically try next spring, and ideally, one or two of them will prove to deliver the effects we need. And then we could organize advocacy around those.

So what comes next for your research in this area?

We’ve been working with a group of 14 districts that are giving us data on which kids they provided tutors to, which kids got double-dose math, and various other things over this past school year. We’ve been working with the NWEA data and hope to have a report out in August laying out the effect sizes that districts got for the interventions they attempted in 2021-22.

As the United States approaches the end of another school year marked irrevocably by COVID, the paper鈥檚 findings are merely the latest to show substantial academic reversals resulting from lengthy school closures, uneven execution of virtual learning, and the effects of countless social and economic upheavals over the last two years.

Study co-author Megan Kuhfeld, a senior research analyst at the nonprofit research group NWEA, declared herself 鈥渟urprised in some ways that the results were a lot bigger鈥� than those following the seminal natural disaster of recent American history. But while the 鈥渃oncentration of disrupted schooling鈥� immediately following Katrina was especially intense, Kuhfeld argued, most affected children were relocated and enrolled in new schools within a matter of months. 

鈥淚f you think about how prolonged the disruptions were, it does make sense to me that we saw larger drops than we did during the hurricane,鈥� she said. 

The study, by Brown University鈥檚 Annenberg Institute for School Reform, examines student performance on NWEA鈥檚 Measures of Academic Progress (MAP) test, which assesses growth in both reading and math. The test is typically administered in participating K-12 schools three times during the school year, and results include demographic information detailing student age, gender and race or ethnicity.

All told, the study鈥檚 sample comprised roughly 5.4 million students in grades 3鈥�8 across a swath of 12,000 elementary and middle schools; those schools represented between 12 percent and 15 percent of all American schools in each grade. To illustrate the learning trajectory of these students during the pandemic, Kuhfeld and her collaborators tracked the scores of same-age test-takers during the fall of 2019, 2020, and 2021.

As expected, the arrow pointed downward over those three academic years. Average math scores dropped by a range of 3.3 points to 4.5 points (sixth graders and eighth graders, respectively)during the period under examination, while reading scores dropped between 1.5 and three points (eighth graders and third graders, respectively).

To put those figures in perspective, the authors followed up with , which compared the negative impact captured on the MAP tests to the positive growth catalyzed by various education policy interventions in separate studies. The pandemic-related hit to math scores among elementary schoolers between 2019 and 2021 was nearly three times as large as the benefit that similar students have received from a specialized course of summer instruction in the subject. Score slippage for middle schoolers was larger in magnitude than the boost that comparable children received from high-dosage tutoring, one of the most effective education interventions that has ever been measured. 

Mike Magee, who recently stepped down as CEO of the education advocacy group Chiefs for Change, called the study 鈥測et another indication that an alarming number of students have fallen far behind during the pandemic.鈥� He called for a range of remedies 鈥� including extending the school day, offering higher-quality instructional materials, and deploying more tutors 鈥� to arrest the downward momentum in achievement.

鈥淐ovid has exacerbated inequities and widened gaps,鈥� Magee wrote in an email. 鈥淚f schools do not have an uncompromising focus on academics, the lasting consequences will be catastrophic 鈥� and Black and brown children, and those who live in poverty, will hurt the most.鈥�

Indeed, the study also detected widening disparities in academic performance between students of different levels of socioeconomic advantage. In just two years, the already sizable gap in reading performance between students enrolled at relatively high-poverty schools (those in which over 75 percent of students are eligible to receive free or reduced-price lunch) and relatively low-poverty schools (those in which fewer than 25 percent are eligible) expanded by 15 percent. The gap in math performance grew by 20 percent.

Longtime education observer Tom Loveless, the former director of Brookings鈥檚 Brown Center on Education Policy, was struck by the sheer toll exacted by COVID鈥檚 disruption to learning, but also the smaller stories within the learning loss.

鈥淲hat’s interesting is that the losses from the pandemic seem to be the opposite of the learning gains registered over the past three decades,鈥� Loveless wrote in an email. 鈥淪ince the early 1990s, greater gains were made by younger kids than older kids, and math gains outpaced gains in reading. Now 鈥� at least in early studies of pandemic effects 鈥� we are seeing greater losses by younger students compared to older ones and a larger setback in math compared to reading.鈥�

The explanation for the subject-specific differences in MAP scores is still unclear, though Loveless added that math performance may simply be 鈥渕ore sensitive to schooling鈥� 鈥� and therefore vulnerable to decline when schools shut down. Evidence for that theory can be found in a curious time lag: Even after the sudden impact of school closures in the spring of 2020, reading scores measured that autumn were still only modestly affected, sinking much further over the course of the 2020-21 school year. But significant declines in math scores were apparent in fall 2020 MAP performance.

Kuhfeld agreed that previous research has shown that math learning 鈥渟eems to be more sensitive to time in school,鈥� which could account for the steeper backsliding in that subject. But she added that the MAP results can鈥檛 capture the lost growth among early readers in the first and second grades, who were just beginning to develop the foundational skills of literacy when COVID hit.

鈥淭hose kids who were first-graders when the pandemic started are now in third grade,鈥� Kuhfeld observed. 鈥淭here’s probably going to be a cumulative effect where kids who missed out on those formative, learning-to-read experiences will probably see even larger negative effects on reading than we’ve observed so far.鈥�

Further research will additionally clarify the extent to which the educational consequences of the pandemic differed among students in different learning environments, she added. One study 鈥� circulated last fall by a team of academics including Brown economist Emily Oster 鈥� has already found that learning loss was much greater in school districts that kept remote schooling for longer. Kuhfeld argued that there was likely to be 鈥渉uge variation鈥� based on the time children spent in physical classrooms.

鈥淚t was probably worse for certain groups that either didn’t have access to in-person instruction, or maybe had access but chose not to go for health or other reasons,鈥� she said. 鈥淭he kind of scary part is knowing that this was the average effect; there were certainly a number of students who had larger drops than what we’re reporting.”

In early March, a pandemic celebrity best known for advocating that schools should move toward a pre-COVID normal wielded her weapon of choice, arguing in The Atlantic that lifting mask mandates for all but the youngest students is 鈥�.鈥�

Emily Oster laid out what she, and many others, understood to be the situation at hand in her opening paragraph: 鈥淎lthough the CDC recently moved to relax COVID guidelines, it continues to recommend universal indoor masking in early-childhood-education programs for those ages 2 and older.鈥澛�

The CDC鈥檚 coronavirus for child care providers, last updated Jan. 28, lists a number of 鈥渒ey takeaways,鈥� including that the agency 鈥渞ecommends universal indoor masking in [early childhood education] programs for those ages 2 years and older, regardless of vaccination status.鈥�

But in a surprising twist, about a week later, the Brown University economist posted an update on her Instagram story.

鈥淎fter my piece in @theatlantic last week, the CDC emailed me to let me know they DO NOT recommend masking for toddlers in areas with low or moderate transmission. Toddlers鈥� masking recommended to align with everyone else,鈥� she wrote. 鈥淭hey are struggling to get the message out so maybe this will help!鈥�

鈥淚 realize that seems a little crazy, but I am telling you that is the email I received from a senior person at the CDC.鈥�

The federal agency has a yellow banner at the top of its that says the CDC鈥檚 latest recommendations 鈥渁lign precautions for educational settings with those for other community settings.鈥�

鈥淭hat banner 鈥� is intended to replace all of the information that is below it in the bullets that say that kids should still be masking,鈥� Oster said in an Instagram video.

In late February, the CDC made major news when it replaced its previous recommendation that all schools require universal masking, stipulating instead that classrooms could now go mask-optional when community COVID rates were low or moderate, the current virus level across most of the country.

But without a vaccine available for those younger than 5, Oster and many others understood the guidance to apply only to K-12 schools, not early child care and pre-K programs. The CDC is 鈥渆asing its recommendations for wearing masks in indoor K-12 settings,鈥� the Los Angeles Times .

But in fact, the guidance was meant to apply to all educational levels, including those under 5.

In a Thursday email to 蜜桃影视, the CDC confirmed that 鈥渞ecommendations for masks in K-12 schools and early care and education (ECE) programs are consistent with recommendations for other community settings.鈥� 

鈥淐hildren ages 2-4 have a lower risk of severe disease from COVID-19 and parents of children in ECE programs as well as ECE staff can make appropriate choices about mask wearing in school settings based on local requirements and their personal levels of risk,鈥� wrote spokesperson Jade Fulce.

She did not explain why it has taken the agency several weeks to update its website, but said they would make the information available 鈥渁s soon as possible.鈥�

To New York City parent Daniela Jampel, whose 4-year-old daughter has continued masking while her older sister goes to school face exposed, the delay is unacceptable.

鈥淚t鈥檚 ridiculous,鈥� she said. 鈥淭he CDC is having trouble updating its website so they reach out to Emily Oster?鈥�

鈥淭heir website on this issue should not be left to interpretation. It should be very clear,鈥� said Jampel, an early advocate for amid remote learning and now an outspoken critic of the city鈥檚 decision to leave masking in place for preschoolers.

Oster agreed that the unconventional communication method underscores the widespread confusion on the issue, but clarified that the CDC did not contact her asking her to spread the word about their policy. Rather, they were correcting what they said was inaccurate information in her Atlantic piece.

鈥淭hey weren’t like, 鈥極h, by the way, it would be great if you could share with people this information,鈥欌�� Oster told 蜜桃影视. 鈥淭hey just said, 鈥楨verybody should already know this.鈥� But I think it’s pretty clear looking at 鈥� how people responded that they have not managed to make that clear.鈥�

Several parents, mostly in blue states like New Jersey, Connecticut and Illinois, responded to Oster鈥檚 update saying that their child care provider was still requiring masks, said the professor.

鈥淚 showed this (post) to my provider,鈥� many parents wrote, and in response were told, 鈥淲ell, if the website still says that masks are required, that鈥檚 not our interpretation of what that banner is.鈥�

鈥淭here is a fair amount of people looking to this guidance and trying to interpret it and the way that it is currently stated is extremely difficult to interpret clearly,鈥� said Oster.

The confusion extended to The Atlantic itself, which did not immediately update Oster鈥檚 original column to reflect the CDC鈥檚 clarified guidance after Oster received the agency鈥檚 email. In a follow-up interview with 蜜桃影视, Oster said she corresponded with her editor, but because the CDC had made no official announcement on how to interpret the vague website, the outlet decided not to alter its story at that time.

鈥淸The fact-checker] read the banner at the top, but then everything below it still said there should be masking,鈥� she said. 鈥淚t went under the radar.鈥�

However, after this story first published and 蜜桃影视 requested comment from The Atlantic, Oster鈥檚 piece was updated Thursday night to reflect the disconnect in the CDC鈥檚 guidance between the banner and the information below it.

Many early childhood education providers nationwide continue to require universal masking for 2- to 4-year olds.

Head Start, a federal school readiness program serving over 800,000 children from low-income families each year, 2-year-olds and up to wear face masks indoors, although in a Jan. 1 ruling, a U.S. district judge on the program鈥檚 rule in 24 states, mostly Republican. In the remaining 26 states, even those that long ago lifted their school mask mandates, participating toddlers are still required to cover up.

New York City Mayor Eric Adams cited hospitalization data when announcing earlier this month that the country’s largest school district was lifting its K-12 mask mandate but keeping the rule for 2- to 4-year-olds.

鈥淲hen you look at those under 5, they were more likely to be hospitalized,鈥� Adams . 鈥淧eople wanted to say, 鈥楲et’s lift it across the board,鈥� but that’s not what the science was showing us.鈥�

Masking in early child care settings is associated with a in program closures due to virus outbreaks, according to a recent study from doctors at Yale University. But the data were collected during the early months of the pandemic before vaccines were available to staff.

And while federal data show that hospitalizations for children under 5 did spike during the Omicron surge, an outsized share of that uptick was driven by newborns not yet 6 months old, who the masking guidance does not apply to anyway.

Meanwhile, COVID cases in Europe are , fueled by a more transmissible Omicron subvariant. Even as infections continue to , many experts warn that the increases across the pond could foreshadow a coming wave in America.

Jampel, despite frustration with the CDC鈥檚 haphazard rollout of its guidance for toddlers, doubts whether more clarity would impact the rules affecting her family. 

鈥淣ew York City schools have done many things that go far beyond what the CDC recommends,鈥� she said. 鈥淚’m not convinced that it鈥檚 the CDC holding us up, and I’m not convinced that a CDC change will mean that our political leaders will take notice and change their policies.鈥�

Neither the Department of Education nor the Department of Health immediately responded to requests for comment.

Steven Barnett, co-director of the National Institute for Early Education Research, said the two key questions on the issue are 鈥淲hat are the health benefits from masking young children?鈥� and 鈥淲hat are the developmental consequences?鈥�

鈥淭he problem with trying to be an expert on this issue is that there is very limited science on which to base conclusions,鈥� he told 蜜桃影视 in an email. 鈥淲ith respect to the health benefits, the known risks to young children from infection are quite small but this is a novel virus with unknown long-term risks.鈥�

鈥淎ll this leads me to think,鈥� he continued, 鈥渢hat masks for young children may be prudent when there is a high rate of community transmission鈥� 鈥� a conclusion that lands him in alignment with the now clarified CDC guidance.

But with all the CDC鈥檚 communication glitches along the way, Oster worries it will impact the public鈥檚 faith in the agency, which has been shaken several times throughout the two-year pandemic.

鈥淭his erodes trust,鈥� she said. 鈥淚f people are trying to trust the CDC, they’re trying to listen to them, when the messaging is confused in this way, or incomplete in this way, it makes people less likely to pay attention to the CDC.鈥� 

It鈥檚 a question that education experts have been asking since the emergence of COVID-19, and one whose answers are gradually becoming clearer. With that 99 percent of students have now returned to classrooms, newly available data are showing how students were affected by spending long stretches of the last two school years at home. And the signs are not good.

Perhaps the most disturbing news yet was found in by the National Bureau of Economic Research, which found that state test scores dropped significantly in both reading and math during the pandemic. In a discovery that will reopen questions about the wisdom of keeping schools closed, economist Emily Oster and her co-authors found that learning loss was far worse in districts that kept classes fully remote, and that declines in reading scores were greater in districts serving predominantly poor and non-white students.

Oster, a Brown University professor and popular author, has in the COVID era as an advocate for school reopenings. One study she co-authored, examining the spread of coronavirus in 250 Massachusetts districts last winter, helped persuade officials at the Centers for Disease Control to reduce the recommended social distancing requirement in schools from six feet to three.

In an interview, Oster said that while the pandemic鈥檚 academic impact was 鈥減robably larger than [she] expected,鈥� the differential effects related to closure policy were not unexpected.

鈥淐ertainly I do not find the direction surprising, or the fact that there was a significant difference across these groups,鈥� she noted.

The study makes use of two huge sources of information. One, the , was launched in September by Oster and her colleagues to track the different learning models (virtual, in-person, or hybrid), enrollment trends, and public health outcomes that prevailed in schools during the 2020-21 school year. 

The other was assembled from the 2021 math and English scores for students in 12 states between the third and eighth grades. The states studied (Colorado, Connecticut, Florida, Massachusetts, Minnesota, Nevada, Ohio, Rhode Island, Virginia, West Virginia, Wisconsin and Wyoming) were chosen because their student participation rates in state tests remained above 50 percent this spring, and they offered at least two years鈥� worth of testing data from the period before the pandemic.

The researchers found that overall student pass rates 鈥� the rate at which students score at or above 鈥減roficient,鈥� however that is designated by the state administering the test 鈥� dropped in all 12 states, though with a wide variation in the size of the declines. In Wyoming, pass rates fell 2.3 points compared with prior years; in Virginia, they plummeted by 31.9 percent. 

What鈥檚 more, the scale of learning loss was far more substantial in areas that kept schools closed longer. 

The team specifically targeted the effects of school closures by dividing all school districts in their sample into three groups: those that offered in-person learning for at least two-thirds of the 2020-21 school year, those that went in-person for less than one-third of the year, and those that fell somewhere in-between. Then they compared changes in test performance among schools that fell into the different categories.

The total effect of a district shifting from 0 percent in-person learning to 100 percent would be to reduce the drop in math pass rates by 10.1 percentage points (or more than two-thirds the average amount they declined during 2020-21, 14.1 points), Oster and her collaborators calculated; the same change would reduce the drop in English pass rates by 3.7 percentage points (more than half the average amount they declined over the same period, 6.3 points). 

The downward movement on achievement was also somewhat linked to student background. By indexing the decline in scores to district demographic information, the authors found that in districts that enrolled over 50 percent African American or Hispanic students, the effect of switching from fully in-person classes to fully remote was associated with a drop in pass rates of 9 percentage points. Meanwhile, in a district enrolling no African American or Hispanic students, that switch only brought about a drop of 4.3 points.

Those disparate trends find support in other research. Recent results from the online i-Ready assessment, administered to over 3 million elementary and middle schoolers across 50 students by Curriculum Associates, showed that students in majority-African American schools have fallen behind those in majority-white schools by a full 12 months of learning during the pandemic. Black students, on average, have enjoyed much less access to in-person classes during that time, studies have demonstrated.

One lingering question is the extent to which the results were influenced by the cross-section of students who sat for tests this spring. With a sizable number of students either opting out of state-required testing or simply leaving public schools entirely, some have wondered whether the students who participated in the exams offer a representative sample from which to draw conclusions. 

Oster said that the high participation rates in states that were selected for the study (all above 80 percent, and most above 90 percent) gave her 鈥渕ore confidence鈥� in the effects she found. If anything, she said, the groups that were underrepresented in spring testing 鈥� disproportionately English learners and special education students 鈥� made it likely that the study was underestimating the damage wrought by the pandemic.

鈥淵ou see pretty consistently across states that there was less participation among English language learners or special ed students. That makes me think that鈥hese numbers could be even larger if we sampled those groups at higher rates also.” 

After just eight days in classrooms, 37 positive coronavirus cases in the 16,000-student district outside Indianapolis had triggered hundreds of student quarantines, forcing young people to miss out on classes and extracurriculars.

Superintendent Jeff Butts knew he had to act fast. The district had begun the year mask optional in late July. But in early August, he stumbled on a solution, hidden in plain sight: The Centers for Disease Control and Prevention had just updated its guidance, exempting students from self-isolation if they and the infected student were properly masked and spaced at least 3 feet apart.

鈥淭hat was my biggest tipping point, quite frankly, when the CDC came out and made that change,鈥� Butts told 蜜桃影视. 鈥淚 realized that if we had all of our children in masks 鈥� I can quarantine fewer children.鈥�

But not everyone got the message. It doesn鈥檛 appear that the guidance trickled down to many other school systems, where , according to a recent survey of 100 districts from the University of Washington鈥檚 Center for Reinventing Public Education. One reason for the disconnect is that CDC made little attempt to billboard the policy shift, which only appears in an on case investigation and tracing updated Aug. 5.

Get stories like this delivered straight to your inbox. Sign up for 蜜桃影视 Newsletter

鈥淚t鈥檚 buried in some appendix to the close contact definition,鈥� Emily Oster, Brown University economist who has tracked schooling through the pandemic, told 蜜桃影视. Under many school systems鈥� quarantine protocols, spending 15 minutes within a six foot radius of an infected individual 鈥� sitting next to them in class, for example 鈥� can force students to stay home for up to two weeks. The new exemption allows schools to bypass that rule in cases where both individuals mask up.

Across the country, as school leaders struggle with quarantine totals that are stretching into the thousands just weeks after schools opened their doors, the new masking exemption to self-isolation guidelines, could help districts sidestep chaotic reopenings amid divisive politics surrounding the use of masks.

In the past week across the country, New Orleans School District after 299 recorded COVID cases. Mississippi has statewide, an official announced. And a district in Texas . Meanwhile, Texas 鈥� among other states like Florida, Arkansas and Arizona 鈥� maintains a ban on mask mandates, though school systems like those in Dallas and Miami are .

The CDC did not respond to 蜜桃影视鈥檚 request for an explanation of why the update wasn鈥檛 publicized more widely. But Oster, the Brown economist, said it鈥檚 possible that when the CDC updated the definition of close contact for quarantining, 鈥渢hey didn鈥檛 realize how important it would be for school guidance,鈥� and thus didn鈥檛 heavily broadcast the change.

At the very least, it鈥檚 clear the hidden clause gives districts a 鈥渉uge incentive to have everybody mask,鈥� Oster said.

As of Aug. 11, all students and faculty in Wayne Township are now required to wear face coverings. Site leaders have told Butts that the district is already seeing fewer quarantines, though the superintendent said he doesn鈥檛 yet have this week鈥檚 numbers.

Wayne Township is not the only locale to pull the trigger on face coverings in response to skyrocketing COVID absences. Elsewhere in Indiana, Greater Clark County Schools adopted a universal masking rule on Aug. 7 after some 70 COVID cases . In Arkansas, the Marion schools superintendent mourned that the state-level ban on mask mandates had caused a nearly in his district. And in Ohio, in an effort to avoid the fate of mass quarantines, Lakota Local Schools outside Cincinnati announced a , just two days before students returned to classrooms.

鈥淏ecause we want to keep our kids in school all year long, just like we did last year, we made a decision this weekend to move to masks,鈥� Superintendent Matt Miller told 蜜桃影视.

Where school systems have the latitude to set their own face covering rules, 鈥渁ll these school districts are probably going to go to masks because there鈥檚 too much COVID right now,鈥� said Dennis Roche, co-founder of the website Burbio, which has tracked school policy through the pandemic.

While exposure to infected individuals often keeps large numbers of students home from school, very few students in isolation actually turn out to contract the virus themselves, Oster noted. Having a rule that allows healthy students to avoid missing class is crucial, she said.

鈥淭he quarantine itself is tremendously disruptive. And so I think that having an off ramp or a way to make it possible for people not to have to quarantine after an exposure is just huge for generating a functioning school system.鈥�

As Delta variant COVID cases continue to surge, allowing students to come to school without masks and spread the virus is inexcusable, said Dan Domenech, executive director of the School Superintendents Association.

鈥淚f you have to now quarantine a student because they’ve been exposed to somebody because nobody was wearing a mask, that’s a problem,鈥� he told 蜜桃影视. 鈥淔rom a logistical point of view, the easiest thing to do is to say everybody needs to wear a mask.鈥�

Despite the potentially large implications for schools鈥� daily operations, there was 鈥渘ot much emphasis鈥� on the CDC鈥檚 policy change, said Domenech 鈥� meaning many districts may still be struggling to catch up.

From a public health perspective, the move aligns with what Phil Chan, medical director for the Rhode Island Department of Health, says are the best practices to prevent the spread of COVID.

鈥淲here we are with our case transmission rates across the country鈥� I think [masking] makes all the sense in the world,鈥� he told 蜜桃影视. It鈥檚 鈥渢he bare minimum we should be doing at this time.鈥�

Still, in his home state, face covering policies in school are 鈥渁ll over the map,鈥� he said, which he fears could spell unnecessary COVID spread and lost learning.

Last week in Georgia, for example, four school districts 鈥� some of which had mask-optional policies 鈥� due to COVID outbreaks.

As summer ends and students return Wednesday to Lakota Local Schools, that鈥檚 precisely the situation that Superintendent Miller hopes to avoid.

鈥淚 think the social emotional pitfall of masking is bad enough, but I think the social emotional pitfall of being at home and learning again from home is probably worse.鈥�