Traditional high-impact tutoring can cost upward of $2,000 per student a year, and staffing is the single biggest constraint. At the same time, shortages of qualified teachers persist, with districts struggling to recruit and retain the educators students need most.

These two crises, a tutoring access gap and a teacher pipeline shortage, are usually treated as separate problems, but they shouldn鈥檛 be. Among the full landscape of education interventions, high-impact tutoring is one of the most consistently effective, evidence-based strategies for accelerating student learning. The results are replicable, offering up a solution to both crises that is currently hiding in plain sight.

Each year, more than 600,000 aspiring teachers are enrolled in educator-preparation programs across the country. Another 600,000 college students are employed through as well as state programs, such as . We can, and must, activate these people as tutors for the students who need them most. To do that, policymakers should act on two fronts.

First, unlock Federal Work-Study dollars for tutoring. The infrastructure already exists. Work-Study employs 600,000 college students annually in federally subsidized campus jobs. Redirecting even a fraction of these positions toward high-quality tutoring would create one of the largest, most cost-effective tutoring workforces in the country without requiring new appropriations.

This is already happening. Step Up Tutoring engages college students paid through Federal Work-Study or College Corps at 40 colleges and universities across 15 states, making it one of the fastest-growing Work-Study鈥損owered tutoring programs in the country. Step Up delivers one-on-one virtual tutoring and mentorship to over 5,000 underserved students annually in more than 40 districts across four states. Its students are outperforming peers by wide margins; an independent evaluation found that students receiving tutoring with Step Up gained two to four additional months of learning in math compared to a control group.

Critically, this model both expands the tutoring workforce and strengthens the educator pipeline. This year, 73% of Step Up’s college and high school-aged tutors reported that they are somewhat to strongly interested in pursuing a career in education, and 82% said their Step Up experience increased that interest. As one tutor shared: “Step Up confirmed my desire to go into teaching. I wasn’t sure before, but working with my student has been the most fulfilling part of my week.”

Second, require tutoring experience as a core component of teacher preparation. Many aspiring teachers enrolled in prep programs don鈥檛 have an opportunity to regularly practice what they learn until a culminating student teaching experience or a year-long residency near the end of their program. Tutoring can be the lab where theory meets practice earlier in their preparation, allowing candidates to begin working directly with students to practice instructional skills and identify and use high-quality instructional materials in real time.

Deans for Impact’s partnerships with nearly 300 prep programs demonstrate that aspiring teachers grow more skilled, confident and effective when they have structured opportunities to engage in on-the-job learning early and often. Through a pilot designed to prepare aspiring-teacher tutors to identify and effectively use high quality materials, there was an average 20-plus percentage-point growth in instructional skills and knowledge among participants. Findings also showed an average overall increase of over 49% in tutors鈥� feelings of preparedness to teach.

When tutoring is embedded into preparation, and not treated as an add-on, aspiring educators build instructional skills earlier, with support, before stepping into the complexity of full-classroom teaching. Districts gain a steadier, stronger pipeline. And states produce teachers who know how to accelerate learning from day one.

There is another reason to be optimistic about the effectiveness of these novice tutors. Increasingly, AI-powered tools can provide real-time instructional guidance, helping tutors decide what to teach, how to explain concepts and how to respond when students struggle. This is not about replacing the human relationship at the center of effective tutoring; it is about ensuring that every willing tutor, regardless of prior experience, can deliver consistent, high-quality instruction.

If we act on these two priorities 鈥� unlocking Work-Study funding and embedding tutoring in teacher preparation 鈥� we can solve two critical problems at once. Students gain the academic support and human relationships they desperately need. And more young adults can build their confidence and skills in teaching from the start. In the process, they establish a habit of service that will shape the rest of their careers.

Despite the sunset of ESSER funds, the federal government has continued to foster momentum by elevating tutoring as a priority in existing and future grant competitions. In December 2025, the U.S. Department of Education awarded $256 million via the to scale tutoring and improve literacy. Also in December, a growing bipartisan, bicameral coalition of Congressional leaders re-introduced the PATHS to Tutor Act to scale local partnerships working to embed tutoring into teacher training. 

But the next step must be bolder: we need a comprehensive, national strategy that integrates tutoring into the fabric of teacher preparation and channels federal dollars toward improving academic outcomes while simultaneously cultivating the next generation of educators. 

For Rachael Garnick, a former first grade teacher, the results were a reminder of how tough it’s been for schools to recover from historic declines in learning since the pandemic. 

鈥淭he literacy scores are still abysmal and we should be displeased,鈥� said Garnick, who heads the Pennsylvania Literacy Coalition. Made up of over 70 organizations, the group has pushed and state officials to fund and implement reading reform.

But despite the discouraging statewide results, she also sees districts, like in northeastern Pennsylvania and the Mohawk Area district, northwest of Pittsburgh, 鈥渢rending in the right direction,鈥� and demonstrating urgency over reading scores. Their attitude, she said, was 鈥渢he opposite of 鈥業f it ain’t broke, don’t fix it.鈥� Instead, 鈥業t鈥檚 broke; we鈥檝e got to fix it.鈥� 鈥� 

on pandemic learning loss from NWEA, an assessment company, captured that combination of frustration and hope over the state of academic recovery. About a third of schools have reached pre-COVID performance levels in reading or math, and just 14% have recovered in both subjects. But even some that were hit the hardest, like high-poverty schools, have made impressive gains.

The report was just the latest collection of results pointing to a long road ahead for most schools. Last year鈥檚 National Assessment of Educational Progress scores showed students in the majority of states losing more ground, but included a few standouts with strong progress, like Louisiana in reading and Alabama in math. And state test scores tell a similar story: few have topped pre-COVID performance.

It鈥檚 not like experts didn鈥檛 predict a slow recovery. 

鈥淚f student performance improvement follows historical prepandemic trends, it could take decades for students to fully catch up,鈥� researchers with McKinsey and Company, a consulting firm, .

Even the nation鈥檚 education chief isn鈥檛 expecting good news soon. 

鈥淚 would like to say that NAEP scores, when they come out again in January 2027, are going to show marked improvement,鈥� Education Secretary Linda McMahon said in a recent K-12 Dive . 鈥淚 don鈥檛 think they are.鈥�

But Dan Goldhaber, director of the Center for Analysis of Longitudinal Data in Education Research, said it鈥檚 important to put NWEA data, and all measures of kids鈥� learning, in context.

鈥淥ne of the reasons that we’re not seeing recovery and that the results aren’t better is because of what was happening in the decade ,鈥� he said. 鈥淭here was a slow degradation of academic achievement.鈥�

Resisters and rebounders

Schools that were able to resist further declines during the pandemic are those that are more likely to be back on track, according to NWEA鈥檚 data, which represents five million students who took the MAP Growth tests through fall 2024. Such schools make up nearly three quarters of the recovered schools.

The Los Angeles-area is one example. 

Before the pandemic, the high-poverty, majority Latino district was already seeing gains on state assessments. When testing resumed in 2022, reading scores held steady. Math scores caught up the following year, and the district has continued to post gains ever since. 

Superintendent Darin Brawley highlighted a mix of academic routines, like a math problem of the day, weekly quizzes and challenging writing assignments, that the district continued despite the disruption of school closures. Teachers were encouraged to dial back their use of smart boards in the classroom and require students to keep math and language arts journals to improve retention. 

鈥淓verything was being done on the smart board and kids weren’t notating anything,鈥� Brawley said. 鈥淐ertain things have to be worked out on paper.鈥�

NWEA data also pointed to what the researchers call 鈥渞ebounder鈥� schools, those that saw significant drops in achievement but have been able to climb their way back. High-poverty schools are among those with impressive gains, but even districts seeing higher-than-ever performance still struggle to close wide achievement gaps.

鈥淲e’ve never had scores this high in English language arts or math,鈥� said Buffy Roberts, associate superintendent of the Charleston County schools in South Carolina. 鈥淚t’s been quite phenomenal.鈥�

She was talking about , which, unlike NWEA and NAEP, aren鈥檛 comparable because states don鈥檛 all measure proficiency the same. But they can still reflect post-COVID trends if states haven鈥檛 changed their tests since 2019. 

South Carolina鈥檚 math test has remained constant. Results show that statewide, scores have nearly recovered. It鈥檚 a trend that NWEA noted as well, explaining that while schools 鈥渓ost significant ground,鈥� in math, many made 鈥渟ubstantial gains afterward.鈥�

In Charleston, 54% of students in grades three through eight met or exceeded expectations in math last year, up from 48% in 2019 and about 10 percentage points higher than the state average. The district also made the Harvard Center for Education Policy Research鈥檚 fully recovered districts in the nation last year.

Roberts pointed to a swift return to in-person instruction and high-dosage tutoring as some of the factors contributing to strong growth. But she said at the outset of the pandemic, leaders 鈥渒new there were some vulnerable groups鈥� that would need 鈥渟tructures and support to mitigate some of that learning loss.鈥�

The district鈥檚 , she explained, provided extra dollars to schools with high-poverty students even when the schools didn鈥檛 qualify for federal Title I funding. The schools used the funds for extra staff to reduce class sizes, incentives to increase attendance and mental health services.

But there鈥檚 still a lot of work to do. In fourth grade math, there鈥檚 a more than 50 percentage point gap between white and Black students, and students from wealthier families outscore students in poverty by 39 percentage points. 

鈥淲e agree that progress must be faster,鈥� the district on Facebook after a conservative community group to the disparities. 

In an analysis of scores, Education Data Center researchers, led by Brown University鈥檚 Emily Oster, were hopeful about continued math recovery in 2026. Of the 32 states that have kept the same math test since before COVID, seven met or exceeded 2019 proficiency rates: Colorado, Georgia, Iowa, Mississippi, Missouri, Rhode Island and Tennessee.

But even if they didn鈥檛, they all made some gains. Despite Pennsylvania鈥檚 decline in reading, for example, its performance in math is less than a percentage point from reaching the 2019 level. 

But the results in reading were less encouraging. Six out of 28 states have met or surpassed pre-pandemic performance. But several others, like Massachusetts, Minnesota and Oregon, remain well off that mark. 

Goldhaber, with CALDER, suggested that states haven鈥檛 seen improvement on tests because parents trust those scores less than the grades kids bring home on report cards and assignments. 

A recent reiterated that point. In a survey of over 2,000 parents, nearly three quarters said they believe grades more than tests when making decisions about their children鈥檚 learning. They鈥檙e also less likely to take action, like seeking out tutoring or other help for their child, when grades are good. 

The problem is that because of grade inflation, which was on the rise even before the pandemic, grades are a less accurate measure of how students are really doing. 

The results of that survey were no surprise to Bibb Hubbard, founder and CEO of Learning Heroes, a nonprofit that focuses on helping parents understand achievement data. She said she鈥檚 been 鈥渟creaming from the rooftops for 10 years鈥� that parents are about their kids鈥� performance. 

鈥淕ood grades do not equal grade level,鈥� she said. 鈥淧arents are deeply engaged, but we can鈥檛 afford to leave them on the sidelines relying on grades alone. The stakes are too high.鈥�

High-quality tutoring has emerged as an important post-pandemic for helping struggling students in public schools. finds that tutoring often results in substantial additional learning gains when delivered during the school day, in small groups with the same tutors and multiple times a week for at least 10 weeks. 

But this often comes with a substantial price tag 鈥� depending on the model and staffing approach, can range from $1,200 to $2,500 per student per year. During the pandemic, many districts relied on federal Elementary and Secondary School Emergency Relief funds to launch or expand tutoring programs, but these have largely expired.

Fortunately, states and school districts have access to other funding streams, which can be combined through 鈥渂lending鈥� and 鈥渂raiding鈥� to cover the costs of tutoring when a single source is insufficient.

Federal Funding 

Though federal funding faced significant uncertainty during the Fiscal Year 2026 budget process, Congress passed a spending package that sustains many of these funding streams, at least for the coming year. 

School districts may use Elementary and Secondary Education Act funds 鈥� federal aid intended to close achievement gaps for low-income students 鈥� for schoolwide or targeted tutoring programs, depending on a school鈥檚 poverty level. ESEA funds, which support the recruitment, training and retention of effective educators, can be used to train staff as tutors and provide stipends to those who take on this additional responsibility. ESEA fund student support, academic enrichment and afterschool programs, which includes tutoring. 

Other federal funds may be used for tutoring programs that aid certain student groups. ESEA funds can be used to train and pay tutors of English learners and Native American and Alaska Native students, respectively. And the can cover the cost of tutoring, instructional materials and tutor professional development when these services are tied to a student鈥檚 Individualized Education Program.

Beyond direct funding, districts can leverage federally supported service and employment programs. AmeriCorps, a national service initiative funded primarily through federal appropriations, has long supplied tutors to low-income districts and schools through full-time programs like City Year. And the federal work-study program helps pay part-time salaries for college undergraduates and graduate students, including those who tutor in K-12 schools. 

Finally, the U.S. Department of Education has听听颈迟蝉听, a regulatory framework for discretionary grant competitions. This emphasis aligns with the department鈥檚 Fiscal Year 2025 literacy grants, which include roughly $89 million recently awarded to seven state education agencies to scale tutoring programs.

State and Local Funding

are playing a pivotal role in sustaining and scaling tutoring programs launched with federal ESSER funds by using funding formulas, policy mandates and infrastructure supports to keep post-pandemic initiatives going. 

Many states have relied on short-term appropriations. Louisiana, for example, paired a K-5 tutoring for low-performing students with an initial appropriation in the 2024-25 school year and another $30 million in 2025-26, though future funding will depend on annual legislative approval. 

And while most state tutoring investments have been one-time commitments, stands out for embedding tutoring in its K-12 funding formula by providing an additional $500 per fourth-grader each year for literacy tutoring.  

Some states have enacted tutoring mandates without funding them. , for example, requires that students in grades 3 to 8 who failed the state assessment the previous year receive tutoring, but districts must use a combination of state, federal and local funds to pay for it. Twenty-four states offer , such as vendor lists or other procurement assistance.

A smaller number of states have statewide programs that recruit, train and place members of their tutors corps in schools. In , this operates through a nonprofit model backed by and philanthropy. The is embedded within the state education department and relies on a combination of expiring federal relief funds and funds from local foundations, nonprofits and city governments.

Districts, cities and counties sometimes offer competitive grants that can fund tutoring, and superintendents can reallocate existing dollars for tutoring through their districts’ annual budget process. In cities with strong mayoral involvement in education, tutoring dollars can be allocated directly through the city budget, as the leaders of ., and have done. 

Higher Education Partnerships

Colleges and universities represent a large, often underused source of potential low-cost tutors. In a 2023 , then-Secretary of Education Miguel A. Cardona encouraged cross-sector partnerships to scale tutoring and highlighted federal work-study as a key resource. The letter noted that when eligible college students tutored school-aged children, the government could cover up to 100% of their wages through federal work-study. That guidance remains in effect, and there have been no subsequent regulatory changes to the program.

Teacher-preparation programs are also well positioned to expand tutoring capacity. At for example, undergraduate education majors are required to serve as tutors in local elementary schools as part of their coursework. Although the tutors are unpaid, their work counts toward their field-placement hours for graduation, giving them more than required by the state. There is no cost for participating public schools. The model enables districts to sustain tutoring at little to no cost, while future teachers gain valuable classroom experience. 

Leveraging Philanthropic and Nonprofit Support 

School districts can also partner with philanthropic and nonprofit organizations. The , launched as a public-private partnership during the Biden administration and now operating independently, offers free and connects districts with vetted providers of staffing, training and financial assistance for tutors, helping to reduce hiring, training and program startup costs.

The upshot is that there are more sources of support for intensive tutoring in public schools than one might think. Tapping them may require education policymakers and practitioners to move money from other programs. But with a large body of research showing increased and meaningful learning gains from high-quality tutoring done during the school day, that shouldn鈥檛 be a difficult decision to make.

But that short-term focus on the averages loses sight of what鈥檚 happening at the tails 鈥� the top performers and the weakest 鈥� and how things have evolved over longer periods of time. 

To zoom out, I worked with Eamonn Fitzmaurice, 社区黑料鈥檚 art and technology director, to build the time-lapse tools below. 

The first one shows you the evolution of 12th grade math scores. This particular test was first administered in 2005 by the National Assessment of Educational Progress. When the 2024 scores came out in September, 社区黑料 wrote about the declines overall and for the lowest-performing students.

The reading scores for 12^th graders are even more depressing. They haven鈥檛 gotten as much attention as the math scores, perhaps because the averages scores haven鈥檛 followed as dramatic of an up-and-down rollercoaster as the math scores have followed.

These graphs show the scores of 12^th graders in math and reading, but it鈥檚 likely that other grades and subjects would show similar patterns. It鈥檚 not just that average scores have declined across a wide range of tests, grades and subjects; we also have a lot more low-performing students than we did in the past. 

While the data presented here are at the national level, any state, district or school leader could see how things are changing in their community. At the classroom or school level, increased variability in student performance makes it harder for teachers to personalize their instruction and for school leaders to design systemwide supports. To get things back on track, policymakers should pay special attention to how their lowest-performing students are faring.

The results could open a new front in the evolving discussion over how to use AI in schools 鈥� and how closely humans must watch it when it鈥檚 interacting with kids.

In a involving 165 British secondary school students, ages 13鈥�15, the ed-tech startup put a small group of expert human tutors in charge of a , or LLM, offered by Google鈥檚 . As it tutored students on math problems via Eedi鈥檚 platform, it drafted replies when students needed help. Before the messages went out, the human tutors got a chance to revise each one to the point where they鈥檇 feel comfortable sending it themselves.

Students didn鈥檛 know whether they were talking to a human or a chatbot, but they had longer conversations, on average, with the 鈥渟upervised鈥� AI/human combination than simply with a human tutor, said Bibi Groot, Eedi鈥檚 chief impact officer. 

In the end, students using the supervised AI tutor performed slightly better than those who chatted online via text with human tutors 鈥� they were able to solve new kinds of problems on subsequent topics successfully 66.2% of the time, compared to 60.7% with human tutors.

The AI, researchers concluded, was 鈥渁 reliable source鈥� of instruction. Human tutors approved about three out of four drafted messages with few to no edits.

Students who got both human and AI tutoring were able to correct misconceptions and offer correct answers over 90% of the time, compared to just 65% of the time when they got a 鈥渟tatic, pre-written鈥� response to their questions.

And the AI only 鈥渉allucinated,鈥� or offered factual errors, 0.1% of the time 鈥� in 3,617 messages, that amounted to just five hallucinations. It didn鈥檛 produce any messages that gave the tutors pause over safety.

The results suggest that 鈥減edagogically fine-tuned鈥� AI could play a role in delivering effective, individualized tutoring at scale, researchers said. Interestingly, students who received support from the AI were more likely to solve new kinds of problems on subsequent topics. 

The key to the AI鈥檚 success, said Groot, was that researchers gave it access to detailed, 鈥渆xtremely personalized鈥� information about what topics students had covered over the previous 20 weeks. That included the topics they鈥檇 struggled with and those they鈥檇 mastered. 

鈥淲e know what topics they’re covering in the next 20 weeks 鈥� we know the curriculum. We know the other students in the classroom. We know whether they’re putting effort into their questions. We know whether they’re watching videos or not 鈥� we know so much about the student without passing any personally identifiable information to the AI.鈥�

That guided the AI鈥檚 strategy about whether students needed an extra push or just more support 鈥� something an 鈥渙ut-of-the-box, vanilla LLM鈥� can鈥檛 do, she said.

鈥淭hey don’t know anything about what the teacher is teaching in the classroom,鈥� Groot said. 鈥淭hey don’t know what misconceptions or what topics the students are struggling with and what they’ve already mastered, so they’re not able to dynamically change how they address the topic, as a human tutor would.鈥�

Human tutors, she said, generally have 鈥渁 really good sense of where the student struggles, because they have some sort of ongoing relation with a student most of the time. An LLM tutor generally doesn’t.鈥�

All the same, even master tutors typically don鈥檛 go into a session knowing a student鈥檚 comprehensive history in a course, including their misconceptions about the material. 鈥淎ll of that is too much information for a human tutor to read up on and deal with while they’re having one conversation鈥� with a student, Groot said.

And they鈥檙e under pressure to respond quickly 鈥渟o that the student is not left waiting. And that’s quite an intensive experience for tutors that leads to a bit of cognitive overload,鈥� she said. The AI doesn’t suffer from that. It needs less than a millisecond to read all of those contexts and come up with that first question.鈥�

Even with their personal connection to students, human tutors can鈥檛 be available 24/7. Groot said Eedi employs about 25 tutors across several time zones who are available to students from 9 a.m. to 10 p.m. every day, but to give students broader access would require hiring 鈥渁n army of tutors,鈥� she said.

The new findings could encourage schools to use AI as a kind of 鈥渇ront line鈥� tutor, with humans intervening when a student is 鈥渄erailing the conversation, or they have such a persistent misconception that the AI can’t deal with it,鈥� said Groot. 鈥淲e think that would be an interesting way to collaborate between the AI and the human, because there is still a really important role for a human tutor. But our human tutors just cannot have conversations with thousands of students at once.鈥�

The new study, published last week on Eedi鈥檚 site and scheduled to appear in a peer-reviewed journal next year, differed in one important way from recent studies that looked at AI tutoring. Researchers at in October 2024 examined AI-assisted human tutoring, in which tutors primarily drove the conversation. But in that case, the AI acted as a kind of assistant, providing suggestions behind the scenes. In the Eedi study, it was the other way around, with AI driving the conversation and humans overseeing it.

Robin Lake, director of the at Arizona State University, said the study is important in and of itself, but also in the context of broader findings elsewhere suggesting that, with proper training and guidance, 鈥淎I can be an incredibly powerful tool 鈥� and certainly has a potential to take tutoring to scale in ways that we’ve never seen before.鈥�

Under controlled circumstances, she said, it鈥檚 also 鈥渙utperforming humans 鈥� that’s really important.鈥�

Lake noted a from Harvard researchers that examined results from 194 undergraduates in a large physics class. They presented identical material in class and via an AI tutor and found that students learned 鈥渟ignificantly more in less time鈥� using the tutor. They also felt more engaged and motivated about the material.

Liz Cohen, vice president of policy for 50CAN and author of the recent book , said the study provides 鈥渧aluable evidence鈥� about new kinds of tutoring. 

But one of its limitations, she said, is that it relied on 13-to-15-year-olds. 鈥淪o immediately I have a lot of questions about if the findings are applicable for younger students, especially using a chat based model,鈥� which may not be a good one for such students.

She also noted that there are many questions around student persistence with AI tutors, including what happens when students get frustrated or aren鈥檛 sufficiently engaged in the work? 

鈥淚 still mostly think that entirely AI tutoring programs are biased towards students who want to do the work or are interested in learning,鈥� Cohen said, 鈥渁nd it鈥檚 pretty easy to see that students who aren鈥檛 bought in or are frustrated are going to give up more readily with an AI tutor.鈥�

She noted that her 12-year-old daughter has experienced problems persisting in an AI-powered math tutoring program. 鈥淪he gets frustrated if she can鈥檛 get the answer and then she doesn鈥檛 want to do it anymore, so I think we need to figure out that piece of it.鈥�

Evidence-based supports exist that can address these challenges 鈥� and more than 400,000 additional adults have helped deliver them in the past three years, . The support includes high-dosage tutoring, which can accelerate learning. Success coaching, which combines academic and social-emotional support and problem solving, improves attendance and achievement. In-school mentoring builds interpersonal relationships that foster school connectedness.

Wraparound support, bringing in community organizations to help with such issues as health and housing, addresses obstacles to well-being. Postsecondary advising gives students clear pathways to adult success and helps them see why full engagement in school matters.    

The challenge schools face in delivering these evidence-based supports more broadly is three-fold. First, they must implement them in ways that align with what the evidence shows drives impact. Second, to be effective, all this must be delivered in the context of supportive, human relationships. This requires 鈥減eople power,鈥� often beyond what existing school staff can provide.

Finally, it takes significant organizational capacity to provide the full range of support to the large number of students who often need them. Schools therefore need strategies to reduce the number of students requiring additional help, along with systems that get the right supports to the right students at the right time.  

Results from a recent nationally representative survey of school principals provide encouraging news about the spread of evidence-based student support. Yet, the findings also offer a sobering reminder of the work that remains to reach all students who need help.  

鈥� a coalition of 250 nonprofits, 200 school districts, and 80 institutes of higher education working to expand evidence-based student supports for all K-12 students 鈥� has partnered with the RAND Corporation to survey the nation鈥檚 principals annually over the past three school years.  

Among the most encouraging findings: about half of the nation鈥檚 public school principals report that their school provides high-intensity tutoring. This grows to two-thirds of schools in high-poverty neighborhoods. These results are aligned with a administered by the U.S. Department of Education. Mentoring and wraparound support can also be found in about half the nation鈥檚 schools.

Nearly all high schools report providing postsecondary guidance. Slightly more than a quarter of schools have success coaches. Over the past three school years, principals report that over 400,000 additional adults have been engaged as tutors, mentors, postsecondary advisors or wraparound support coordinators in their schools.

Clearly, evidence-based student supports have expanded substantially since the pandemic. 

Moreover, half of principals report partnering with a local college or a nonprofit to provide some of these programs. Connections with community organizations that were frayed by the pandemic have been rebuilt and strengthened, bringing more adults into schools to provide critical support for students.

Finally, more principals report the use of student success systems to monitor student progress on key indicators, enabling more proactive and strategic action.  

Despite the positive news, the plurality of principals reports that only some to a few of their students who need help are receiving it. Only 20% to 30% of principals report that most or all students get the support they need. Principals do not see student needs decreasing, four years from the height of the pandemic: Less than 10% reported that fewer students needed support in 2024-25, than in prior years.

When asked what stands in the way of more students receiving support, principals report both supply and demand constraints. Half cited funding as an issue, and a similar share said staffing was a challenge. A third reported that finding enough time in the school schedule was a barrier. Smaller but significant numbers cited lack of student interest, parental reluctance and limited awareness of the support available. 

 A number of schools surveyed have overcome these challenges and scaled evidence-based approaches. About 20% of schools could be described as 鈥渇ull-student support schools,鈥� where principals report providing high-intensity tutoring, mentoring or success coaching, and wraparound support. Around one in five principals report providing high-intensity tutoring or mentoring to more than 30% of their students, at least three times as many as typically received this support pre-pandemic.

This shows that significant numbers of schools, including those who serve high-need student populations, have figured out how to solve the challenges associated with providing a wide range of help to large numbers of students.  

Some schools are trying new approaches, including using Federal Work-Study dollars to support eligible college students working in K-12 schools as tutors and mentors, as well as developing pathways from tutoring or success coaching into teaching careers.

Others are tapping one of the most underutilized and most affordable sources of people power: high school students themselves. Good models exist and in a forthcoming survey from TeenVoice the majority of high school students said they would use peer supports, if offered, and would also be interested in providing them. There are also ongoing efforts to identify how high-intensity tutoring, mentoring and postsecondary advising can be delivered both online and in person. Finally, efforts are underway to broadly expand student success systems.  

For the kids to be alright, we need to provide schools where they want to be, schools where they receive high-quality instruction, and schools where they receive the support they need to attend regularly, focus in class, complete their schoolwork and thrive. We are making progress but still need to scale what has been proven to work. 

Since last year, more districts have been publicly sharing information about their AI strategies, and these are increasingly sophisticated and varied. While only a few are using AI to fully rethink instruction or tackle longstanding challenges, the total number of districts in our database of early adopters has nearly doubled in a year, from 40 to 79, with the most growth happening in urban areas. Below are initial findings from the 2025-26 early adopter cohort.

More early adopters, including more urban districts, are piloting systemwide AI strategies

The number of districts offering teacher professional development is up to 86% this year from 63% last year. More districts are providing guidance about AI use (78% this year versus 65% last year), and more are giving AI-powered apps and tools to educators (77% this year versus 70% last year) and to students (63% versus 58%).

Several large, urban districts are shifting from isolated, piecemeal implementation to more transparent, complex and strategic approaches that go beyond tool testing or teacher literacy. 

For example, Denver Public Schools recently published a that includes a continuous-learning framework and guidance. The district鈥檚 vision moves beyond  simply providing instructions on using popular apps to highlighting AI鈥檚 potential to improve data-driven decision-making, real-time coaching and personalized learning. It also outlines ways AI can aid students with disabilities and multilingual learners. The district about AI preferences and usage to develop its handbook. 

Charlotte-Meckenberg Schools in North Carolina, which released its and in spring 2025, takes a different approach. The district will provide training on and access only to specific, targeted AI tools it approves and continue a districtwide ban on others, including ChatGPT. It has designated 30 “AI Champion Schools” to explore the technology and share what they have learned. 

Early adopters are also focusing more on students

As AI literacy efforts for teachers increase, more districts (63%) are also prioritizing student-centered resources and tools. 

Several are using tools created by, or customized for, their students. For example, at , students developed an AI assistant-powered platform in their Java Programming I course to answer questions on topics ranging from homework to current events. The skills students strengthened while working on this project, like collaboration and problem solving, are among those outlined in the district’s . Students also helped design a . 

While more districts are bringing AI to students, only a few are sharing information about shifts in learning standards, which helps build community buy-in and trust. is an exception, announcing it would implement a new curriculum on how to use artificial intelligence, with age-appropriate lessons required by the school board. The district also said it would offer electives on AI for seventh and eighth graders.

A very small number of districts are systematically engaging students in building AI strategy. One example is in California, which asked high schoolers to help shape policies. Selected teens serve as tech interns who facilitate dialogue among students, teachers and school leaders about AI and developed a chatbot that contributes to draft artificial intelligence policies for the district. 

Reimaginers continue to push the edge of what鈥檚 possible

Last year, we identified a small cohort of , districts that include AI strategies in broader plans for change. Even a year later, only a few districts qualify as reimaginers, but their progress is promising. They are diving deeper in 2025-26, piloting AI to transform schools or student learning. 

In fall 2025, ASU Prep opened a in collaboration with the . It uses a mastery-focused, project-based learning model that blends personalized learning with seminars modeled on the Socratic method, where teachers engage students in dialogue and encourage them to develop and defend a point of view. 

Bullitt County Public Schools is piloting a suite of custom-made, AI-enabled coaching tools to help teachers and school staff transition to competency-based learning. Developed in partnership with and built with , some provide 1:1 coaching for teachers, instructional coaches, school leaders and central office teams. Others help educators and leaders design learning experiences and reflect on whether their instructional practices and other initiatives are helping students. 

Districts must advance coherent AI strategies for teaching and learning

While it’s encouraging to see more districts embracing AI, adoption remains fragmented, and district leaders and teachers are still overwhelmed. Early adopters must keep their focus on systemic transformation, not isolated innovation. 

Strategies to do this include:

• running more short pilots, analyzing the results immediately and adopting what works, to keep up with emerging tools and technology shifts;
  
• testing ways to use AI to streamline administrative or operational tasks;
  
• aligning AI solutions to a broader vision of teaching and learning;
  
• committing to helping district leaders, educators and students understand the risks and potential benefits of using AI;
  
• including students in designing strategy and solutions. 

The emerging data on early adopters signals that more districts and schools are willing to embrace AI鈥檚 potential to transform education. At the same time, students have yet to recover from , reminding education leaders what is at stake. Without a coherent vision or systemic strategy for learning, districts risk adopting AI in fragmented ways that widen student achievement gaps and yield short-term gains rather than genuine transformation.

These scores are not just numbers; they signal lost earning potential for today鈥檚 children and weakened competitiveness for tomorrow鈥檚 workforce. The pandemic deepened the decline, but students were already behind. Without action, the cost will be measured in lost opportunity and billions in economic losses.

from Stanford shows losses in student achievement before and after the pandemic equal those during the pandemic, and that the losses are continuing. The study found restoring student achievement to 2013 levels would raise the lifetime earnings of today鈥檚 average student by an estimated 8% 鈥� producing dramatic and sustained gains for our nation鈥檚 economy.

For our kids to be more financially successful as they age into the workforce, schools have to reverse NAEP declines over the past decade. There鈥檚 no time to spare.

This year鈥檚 NAEP proficiency results for public school students show reading scores have reached their , with only 29% of eighth graders and 30% of fourth graders achieving a proficient score. While the slide in has slowed, scores still remain below pre-pandemic levels, and the performance gap between high- and low-performing students has .

Lower performances by today鈥檚 students mean a down the road; proficiency in literacy and numeracy has been linked to several , including more fruitful college opportunities and higher wage jobs.

The research from Stanford estimates learning loss over the past decade has cost our country over . This translates into having an average of 6% higher GDP every year for the rest of this century if students were still at 2013 NAEP proficiency levels.

At the individual level, the average current student can expect to have a lifetime income that is 8% below that of a 2013 graduate. Because disadvantaged students have suffered deeper learning losses, their incomes can be expected to fall by over 10%.   

For our students to earn more 鈥� and be able to compete with their peers worldwide 鈥� educators can鈥檛 leave their outcomes to guesswork. Schools need to ensure students are learning the fundamentals using evidence-backed methods 鈥� and constantly and consistently measure their progress using clear, objective standards.

Some states made noteworthy progress on NAEP this year: Mississippi, Louisiana and Tennessee. Each has a track record of high expectations and strong accountability.

These states use an that puts reading and math achievement front and center. They measure what matters 鈥� proficiency and growth 鈥� and they report results in a way families and educators can understand. Transparency and rigor are fueling their progress.

It has been very difficult to implement effective large-scale reforms, but we now have examples of getting strong increases in literacy and mathematical proficiency when evidence-based policy solutions are implemented faithfully. For example, states are seeing academic growth using the following three approaches.

First, states need to invest in effective personnel. They can do this by and by supporting strong teaching through professional development in evidence-based practices such as use of high-quality instructional materials and assessment data to inform instruction. Further, hiring of math and has shown success.

Using data from and , which are short assessments to flag struggling students early, has helped ensure schools are using necessary interventions with high quality instructional materials. While many successful states mandate the use of screeners, others can incentivize districts to use them by providing the materials for free.

Finally, Alabama has shown that it is possible to begin turning around the math problem. Two years after passing the , Alabama has returned to for fourth grade NAEP math, jumping from last in the nation in 2019 to 31^st this year. This comprehensive math law includes such as elementary school math coaches; increasing the amount of math instruction per day to 60+ minutes; and the adoption of high-quality instructional materials.

Students aren鈥檛 going to catch up if states don鈥檛 make their progress a priority. Some states are leading the way, but more policymakers need to focus on improving student outcomes using tested methods that raise the bar and measure progress. The nation鈥檚 collective economic future depends on rewarding effective schools and reversing the achievement slides of the past dozen years.

The tells a sobering story that should fundamentally change how the country thinks about education policy.

According to the National Assessment of Educational Progress, American students reached their peak achievement in 2013. Since then, they’ve been sliding backward. The pandemic accelerated that decline, but it was far from the whole story.

Since 2013, students have lost nearly three-tenths of a standard deviation in combined math and reading scores 鈥� the equivalent of more than a year鈥檚 worth of education. Only about half of that decline happened during the pandemic years. The rest occurred before COVID-19 existed or since schools reopened.

In reading, less than a quarter of the total drop occurred during the pandemic period. The rest happened before and after, including significant losses from 2022 to 2024. 

Only California, Hawaii, Mississippi and Washington, D.C., are performing better today than at the start of this century.

What does this mean for today鈥檚 students? Based on research linking educational achievement to earnings, the average student in school today will earn about 8% less over his or her lifetime compared with students who attended school in 2013. For some states, that means lifetime earnings losses approaching 14%.

And the nation will suffer from a less skilled workforce. If it could return to 2013 achievement levels, the nation’s GDP would be 6% higher every year for the rest of the century. The present value of what America is losing equals roughly three times the entire current economy.

But policy responses have focused almost entirely on pandemic recovery 鈥� summer school, tutoring and extended learning time. These miss the bigger picture. Even if these policies worked perfectly and returned students to 2019 levels, America would still not be addressing the decade of decline that predated COVID.

More troubling: Despite federal spending of $190 billion on recovery efforts, student achievement has continued to decline since 2022. This suggests, among other things, that the declines from 2019 to 2022 cannot all be attributed to the pandemic, but instead indicate more fundamental problems facing America’s schools.

Perhaps it is time to admit that just adding new programs and mandates on top of a misaligned system is not working. Calls to reform the system date back at least 40 years, when “A Nation at Risk” warned about failing schools. The response has been broad and consistent. We’ve tried everything: stricter graduation requirements, smaller classes, better teacher pay, new curricula, technology, charter schools, consequential accountability systems and billions more in funding.

The result? After four decades of effort, 13-year-olds taking reading tests today perform no better than their counterparts in 1975. Math scores have also fallen back to where they were decades ago. And longstanding achievement gaps have failed to close.

This consistent failure reveals something important: We keep trying to fix a system that’s fundamentally resistant to change. Even when individual districts implement successful reforms that improve student achievement, other districts don’t copy them. When Washington, D.C., and Dallas introduced performance-based teacher systems that led to significant gains, such successful innovations didn’t spread.

This isn’t because educators don’t want students to succeed. It’s because the system doesn’t align incentives with student achievement. Schools and personnel are neither systematically rewarded for improving outcomes nor held accountable for poor performance in ways that drive meaningful change.

The nation must transition from input-based policies 鈥� mandating specific programs or spending 鈥� to outcome-based accountability, focusing on whether students actually learn. On results rather than compliance.

Instead of treating all schools the same regardless of performance, policymakers should pay attention to how well they are performing. High-performing systems should get operational flexibility. Low-performing systems need structured intervention.

The has developed a thoughtful framework for this kind of systemic change, emphasizing student outcomes, local flexibility and state oversight based on performance. Their model recognizes that because schooling is inherently local, the federal and state governments should support rather than micromanage, creating conditions for innovation while ensuring accountability.

America’s economy has thrived despite educational shortcomings, partly because of the strengths of its underlying economic system and partly because the nation attracts skilled immigrants. Foreign-born workers are not only central to many STEM fields but also leaders of some of the largest firms. For example, the current CEOs of Microsoft, IBM, Alphabet, Tesla, NVIDIA and Adobe are all immigrants to the U.S. But America cannot count on always being the location of choice for innovative foreign workers. Some U.S.-trained graduate students from China and India are electing to return home rather than moving to Silicon Valley, and others who would like to stay are having trouble obtaining work permits.

This raises the distinct possibility that the U.S. will have to rely completely on workers who are born and educated here. Unfortunately, on recent international assessments, the United States ranked 34th in mathematics 鈥� below the average for developed nations. This placed U.S. 15-year-olds slightly ahead of the Slovak Republic but behind Malta.

Students in school today will compete globally with peers who currently outperform them. The economic consequences won’t be theoretical 鈥� they will be these students’ daily reality in the job market.

The nation has tried incremental reform for over 40 years. It’s spent hundreds of billions. And still, American students are learning less today than they did decades ago.

It’s time to stop adding programs to a resistant system and start building one designed for success. The nation can continue down the path of gradual decline, or it can build an education system that prioritizes what matters most: ensuring that every student learns. As the federal government retreats from mandates and regulations, it is time for the states also to reconsider their role.

The choice is ours. The question is whether we’re ready to choose fundamental change over comfortable but ineffective incrementalism.

States that joined a national pledge led by three high-profile education advocacy and research groups to cut chronic absenteeism in half over five years fared better. The 16 states and Washington, D.C. posted results 鈥渟ubstantially above the average rate鈥� of decline, though exact numbers are not yet available, said Nat Malkus, deputy director of education policy studies at the American Enterprise Institute, one of the trio.

The national chronic absenteeism average dropped from 28.5% in 2022 to 25.4% in 2023, and fell an additional two points to 23.5% in 2024. Virginia, which is among the 16 participating states, cut its chronic absenteeism by 4.4 percentage points, year over year, to 15.7%, as of spring 2024.

Speaking of the states collectively, Malkus told 社区黑料, 鈥淭hat鈥檚 good but it鈥檚 not as good as we need it to be. I think it points to the need for sustained pressure and a sustained campaign to bring absence rates down and to bring more students back to consistent attendance.鈥�

Last July, AEI and EdTrust, right-and left-leaning think tanks, respectively, and the national nonprofit Attendance Works joined forces to launch The 50% Challenge. This week, the organizations hosted in Washington, D.C., to report on their progress, re-up the call to action and hear insights from state, district and community partners on how they are improving student attendance and engagement.

With California and Georgia recently joining, the 16 states and D.C. who signed on to the pledge account for more than a third of all students nationally. While Malkus doesn鈥檛 necessarily attribute their better results to the pledge itself, he noted that their participation shows a willingness to commit to the cause and be publicly accountable for their results. 

鈥淚 will hold their feet to the fire on this goal,鈥� he added during his opening remarks in D.C.

While felt most acutely by students of color and those in poorer districts, the spike in chronic absenteeism 鈥� students missing more than 10% of school days a year 鈥� regardless of size, racial breakdown or income. Chronic absenteeism from 13.4% in 2017 to 28.5% in 2022 before beginning to drop in 2023.

Only about one-third of students nationally are in districts that are on pace to cut 2022 absenteeism in half by 2027, according to an , and rates improved more slowly in 2024 than they did in 2023, 鈥渞aising the very real possibility that absenteeism rates might never return to pre-pandemic levels.

Research has shown that students with high rates of absenteeism are more likely to fall behind academically and are at a greater risk of dropping out of school. About 8% of all learning loss from the pandemic is attributed just to chronic absenteeism, according to soon-to-be-released AEI research.

The continued disproportionate impacts of chronic absenteeism were confirmed by recent , which found that in roughly half of urban school districts, more than 30% of students were chronically absent 鈥� a far higher share of students than in rural or suburban school districts.

RAND also found that the most commonly reported reason for missing school was sickness and one-quarter of kids did not think that being chronically absent was a problem.

SchoolStatus, a private company that works with districts to reduce chronic absenteeism, also released  for some 1.3 million K-12 students across 172 districts in nine states. Districts using proactive interventions, the company reports, drove down chronic absenteeism rates from 21.9% in 2023鈥�24 to 20.9% in 2024鈥�25.

At this week鈥檚 event, numerous experts across two panels emphasized the importance of a tiered approach to confronting the issue, which has resisted various remedies. Schools must build enough trust and buy-in with kids and their families that they are willing to share why they are absent in the first place. Once those root causes are identified, it is up to school, district and state leaders to work to remove the barriers.

And while data monitoring must play a significant role, it should be done in a way that is inclusive of families.

鈥淲e need to analyze data with families, not at them,鈥� said Augustus Mays, EdTrust鈥檚 vice president of partnerships and engagement.

It鈥檚 imperative to understand the individual child beyond the number they represent and to design attendance plans and strategies with families so they feel supported rather than chastised.

鈥淚t’s around choosing belonging over punitive punishment,鈥� Mays added.

One major and common mistake schools make is “accountability without relationships,鈥� said Sonja Brookins Santelises, the superintendent of Baltimore City Public Schools.

鈥淵ou can鈥檛 鈥榩ull people up鈥� if you don鈥檛 have enough knowledge of what they鈥檙e really going through,鈥� she said.

Panelists were transparent that all this would require immense funding, staff and community partnerships.

Virginia achieved its noteworthy drop in chronic absenteeism after launching a $418 million education initiative in the fall of 2023, in part after seeing their attendance data sink, with about 1 in 5 students chronically missing school. At least 10% of those funds are earmarked to prioritize attendance solutions in particular, according to panelist Emily Anne Gullickson, the superintendent of public instruction for the Virginia Department of Education.听

These strategies are far-reaching, she noted: Because parents had been told throughout the pandemic to keep their kids home at the slightest sign of illness, schools partnered with pediatricians and school nurses to help counter the no-longer-necessary 鈥渟tay home鈥� narrative.

Gullickson said she also broke down bureaucratic silos, connecting transportation directors and attendance directors, after realizing the role that transit played in chronic absenteeism. The state now has second chance buses as well as walking and led by parents or teachers along a fixed route, who pick up students along the way.

And they are 鈥渙n a mission to move away from seat time and really deliver more flexibility on where, when and how kids are learning,鈥� she said. 

鈥淭his isn鈥檛 one strategy. It鈥檚 a set of strategies,鈥� said Attendance Works founder and executive director Hedy Chang, who moderated the panel.

In Connecticut, state leaders have launched the , a research-based model that sends trained support staff to families鈥� homes to build relationships and better understand why their kids are missing school. 

A recent study confirmed that six months after the program鈥檚 first home visits, attendance rates improved by approximately 10 percentage points for K-8 students, and nearly 16 percentage points for high schoolers, said Charlene Russell-Tucker, the commissioner of the Connecticut State Department of Education.

Schools must also work to motivate kids to want to show up in the first place, panelists said, making it a meaningful place that students believe will support and help them in the long run. The only way to do this is to start with student and family feedback, said Brookins, the Baltimore schools chief.

During the pandemic many parents saw up-close for the first time what their kids鈥� classrooms and teacher interactions looked like, 鈥渁nd I don鈥檛 think a lot of folks liked what they saw for a variety of different reasons,鈥� Brookins said.

鈥淚 think it opened up boxes of questions that we 鈥� as the education establishment 鈥� were unprepared to answer,鈥� she added. But chronic absenteeism cannot be successfully fought without engaging in those uncomfortable conversations.

Disclosure: The Bill & Melinda Gates Foundation provides financial support to EdTrust and 社区黑料.

In eighth grade reading, the percentage of students on grade level or above in those schools grew at three percentage points over last year 鈥� from 36% to 39%. In math, the percentage of fourth graders on track in majority-Black schools grew from 36% to almost 40%, the latest i-Ready assessments from Curriculum Associates found.

Those are 鈥渂right spots鈥� in a snapshot that otherwise shows recovery has remained stagnant five years following the pandemic, said Kristen Huff, head of measurement at Curriculum Associates. 

Black students 鈥渉ad a bigger dip, especially in the early grades, so they have more room to catch up,鈥� she said. But generally, performance has plateaued and there鈥檚 still a long way to go to reach 2018-19 levels. 鈥淚 think we have to hold ourselves accountable to at least that bar, but that’s not the end goal.鈥� 

The 2024-25 data, shared exclusively with 社区黑料, represents almost 12 million K-8 students in reading and more than 13 million in math who took the i-Ready tests during the last school year. Unlike the National Assessment of Educational Progress, the i-Ready adjusts questions to students鈥� level. The prompts are more advanced if kids are working above the benchmark and easier if they鈥檙e below, offering teachers a view, Huff said, of how much progress students need to make to catch up. Nearly half of fifth graders, for example, are on grade level in reading, while 29% are two grade levels or more below, the results show. The picture is similar in math, with 53% on target and 20% far behind.

While students are learning, they鈥檙e not mastering as much material as their peers did before COVID. Learning loss is more pronounced in the younger grades, confirming that even those students who were too young to attend school were affected by the disruption. Multiple studies have shown that economic hardship and fewer opportunities to socialize left less prepared for school. In reading, 60% of first graders 鈥� those who were toddlers during the early years of the pandemic 鈥� are on grade level. That鈥檚 down from 68% in 2018-19.

鈥楽light improvement鈥�

Majority-Black schools, however, were well behind majority-white schools before COVID 鈥� by roughly 20 percentage points. Their scores also saw a steeper drop off after the pandemic. 

About a year after the pandemic, McKinsey &Company, a consulting firm, used i-Ready data that students in majority-Black schools were a full year behind those in predominantly white schools, an increase of three months over the prior achievement gap.

鈥淏lack students were often at the lowest achievement levels in many districts,鈥� said Kareem Weaver, co-founder of Fulcrum, an Oakland, Calif.-based nonprofit that provides literacy expertise to school districts. 鈥淚t makes you wonder what was happening before for students to be at a level where even slight improvement is considered noteworthy.鈥�

If students don鈥檛 acquire strong reading skills and basic math facts in elementary school, they won鈥檛 be able to keep up with more challenging assignments, said Ameenah Poole, who worked as a high school administrator in East Orange, New Jersey, until 2022. Her former colleagues, she said, often wondered why students came to them as struggling readers and lacking proficiency in math. 

鈥淭hese foundational skills are paramount,鈥� said Poole, now principal of Ecole Toussaint Louverture Elementary in the district. 

In a school already not meeting expectations under the state鈥檚 accountability system, the pandemic just put kids further behind. Many parents in the 84% Black school have jobs in the service industry. Some are nurses, one drives an Amazon truck, Poole said, and most parents didn鈥檛 work from home when schools went remote.

A lot of students didn鈥檛 even log in to class, and rebuilding attendance routines has been slow and sometimes futile, she said.

鈥淭he culture during the pandemic and post-pandemic [was] that school was an option,鈥� she said. 鈥淲e say, 鈥業f you miss a day, you miss a lot.鈥� Students have to be here in order for us to teach them.鈥�

On New Jersey鈥檚 state test, 19% of third graders met the standards in reading in 2022. That鈥檚 the same year the district began using i-Ready. Students work on skills like phonics and vocabulary or measurement and geometry in 40-minute blocks every week. 

At first, the extra instruction didn鈥檛 translate into higher scores. In fourth grade, the percentage of students reaching the proficient level actually fell to 11%. But when those same students were fifth graders in 2024, Poole began to see the payoff. Thirty-five percent met or exceeded the goal. 

That still means the majority of students are working below grade level, which the i-Ready data also shows. 

Student learning is 鈥渕oving in the right direction,鈥� said Huff with Curriculum Associates, 鈥渂ut it’s not accelerating fast enough.鈥�

The way the i-Ready results are reported, however, could be hiding some improvement, suggested Dan Goldhaber, director of the Center for Analysis of Longitudinal Data in Education Research. 

Identifying the total percentage above and below the threshold doesn鈥檛 capture those students who may have moved up a level or two over time. Districts are 鈥渇ar from full recovery,鈥� he wrote in an from 28 states. But he concluded that $190 billion in COVID relief, the largest-ever one-time infusion of federal funds for schools, contributed to a significant increase in math performance during the 2022-23 school year. 

Mark Sullivan, superintendent of the Birmingham City schools in Alabama, saw evidence of that in his district.

鈥淚 told the teachers, 鈥榊ou [will] have to teach like you’ve never taught before,鈥� meaning that we had to make up multiple grades within a year because of unfinished learning,鈥� he said. 

A 2024 Curriculum Associates showed that Birmingham, where 89% of students are Black and 86% qualify for free or reduced-price lunch, outpaced the state in math recovery after the pandemic. The Educational Opportunity Project at Stanford University the district for the same reason.

Leaders rearranged the calendar so that at the end of every nine-week session, students had a week off. But teachers provided optional instruction during that open week. About 7,000 students participated 鈥渨hen they didn鈥檛 have to come to school,鈥� Sullivan said. 鈥淲e鈥檙e seeing the fruits of that.鈥�

Now, as the department boxes up its pens and pencils, key responsibilities are scheduled for scattering to, from Health and Human Services to the. Amid this reshuffling of federal furniture, however, one critical detail has been neglected: Who will track student outcomes?

Educators and policymakers from across the political spectrum caution that data systems such as IPEDS and assessments like National Assessment of Educational Progress (NAEP) are essential barometers of America鈥檚 educational health. Even Project 2025 cedes that the federal government should maintain 鈥溾�� abilities in education. As funding shifts to other agencies, it鈥檚 urgent that measurement moves with the money. Efficiency without evidence is mere guesswork, and America’s children deserve better than educated guesses.

Consider NAEP, which is written into federal law as a critical audit of whether our educational systems are serving students well. In February, the Trump administrationabruptly cancelled the test for 17-year-olds, bucking decades of uniform collection. This decision came just days after officials had assured the public that NAEP would not be impacted by budget cuts. 

A few weeks later, the department pulled the same stunt with money: A notice suddenly ordered states to liquidate whatever was left of their pandemic-relief funds that very day, freezing nearly $3 billion that districts had already earmarked. 

For now, pandemic relief funds , and for now, most of NAEP testing is back on schedule for 2026, with some conservative experts even proposing it be expanded to an annual schedule from a biennial schedule. But this cancel-then-revive whiplash has already forced , burning scarce time and taxpayer dollars that could have gone toward students.

These disruptions are part of a broader pattern: a systematic dismantling of the nation鈥檚 education data infrastructure. Many of the staff and contractors who compile crucial K鈥�12 databases such as the Common Core of Data (CCD) have been terminated. The National Center for Education Statistics (NCES), which oversees these efforts, axed nearly everyone. Additionally, The Institute for Educational Sciences (IES) has been slashed to a staff that can鈥檛 鈥�.鈥�

While the administration promises that other agencies will disperse federal education dollars, we have received no such promise for the measurements needed to track the impact of these investments; currently, they seem to be abandoned.

All of this is happening at the worst possible time. America鈥檚 students are in an ongoing academic crisis, one that we can only grasp because of national data. After the disruptions of the COVID-19 pandemic, test scores plummeted. According to NAEP results released last year, the average 13-year-old in the U.S. lost ground in both reading and math: scores fell by 4 points, and scores by 9 points compared to just before the pandemic. They are now at their lowest.

This is a moment when we need more educational measurement and transparency, not less. National exams like NAEP are 鈥渢he canary in the coal mine,鈥� alerting us to academic problems while there鈥檚 still time to act. Without a nationwide lens like NAEP, data warehouses as strong as the CCD and IPEDS, and research agencies dedicated to translating these results such as IES, such alarming trends couldn鈥檛 be convincingly demonstrated.

Likewise, the troves of school data that the Education Department aggregates allow educators and researchers to identify where progress is (or isn鈥檛) happening. If those data collections halt, we鈥檒l be stuck funding systems we can no longer evaluate, rewarding failure as easily as success. 

Fragmented, inconsistent state reporting is a recipe for lost information. We鈥檒l return to an era of patchwork statistics that can鈥檛 be compared or evaluated nationally, and the losers will be the students whose struggles are rendered invisible.

Other federal agencies certainly have the capacity to oversee large-scale, results-oriented programs. For decades, HHS has successfully administered, the federal early childhood education initiative. Precisely because it measures outcomes rigorously, Head Start has demonstrably improved for its participants, even now as it faces sweeping budget cuts. Similarly, HHS鈥檚 Children鈥檚 Health Insurance Program (CHIP) has systematically tracked child insurance rates, contributing to a sharp decline in the rate of  uninsured children from. These programs succeeded because funding came paired with robust outcome measurement.

If such agencies are to steward America’s K鈥�12 funding effectively, they must establish or absorb a dedicated educational data arm. One approach could be transferring NCES, preserving essential national assessments like NAEP and comprehensive data collections. Another might be forming an inter-agency task force specifically for education metrics. Even strong advocates of federalism recognize that comparable data across states is indispensable for judging progress; without it, monitoring becomesuneven and incomplete.

Dismantling the Department of Education is billed as streamlining governance. But true efficiency requires more dedicated outcome tracking, not less. When schools outcomes in the past, progress stalled and achievement gaps widened unnoticed. 

Federal data collection has exposed inequities affecting English learners and students with disabilities, problems states were compelled to address once revealed. Removing these accountability tools risks concealing rather than solving such critical issues.

Fiscal conservatives pride themselves on demanding receipts, progressives on demanding equity. Without national metrics, both lose their yardstick, and students lose most of all.

鈥淏ig picture, AI is not very good at math,鈥� said Alex Kotran, co-founder and CEO of . 鈥淟anguage models just predict the next word. You get mixed results using language models to do math. It’s not yet mature enough to where it can be trusted to be scaled.鈥�

And even if it were to improve, critics worry it might hurt kids鈥� ability to try 鈥� and fail 鈥� on their own. Much would be lost, Kotran said, if 鈥渨e get rid of productive struggle and we build this instinct where the first thing you do is go to AI for help.鈥�

But students in the United States and the United Kingdom have a different view. found 56% believe AI could go a long way in reducing math anxiety. 

Fifteen percent of the 1,500 16- to 18-year-old students surveyed said they had already experienced this relief themselves and slightly more than 1 in 5 said their math scores improved because of the technology. 

The survey also included . Sixty-one percent suggested students view AI as 鈥渁 mentor or study partner rather than a crutch鈥�, while nearly half 鈥渟ee value for students in using AI for help with the process of learning math concepts, rather than to give answers.鈥� 

Nicole Paxton, principal of Mountain Vista Community School in Colorado Springs, said her teachers use AI in many ways. Tools like MagicSchoolAI analyze student responses to math prompts, with AI generating 鈥渟pecific, standards-aligned feedback for each student, focusing on their reasoning, accuracy, and math vocabulary.鈥� 

Paxton said the tool highlights strengths and misconceptions, 鈥渨hich helps teachers give timely and targeted next steps.鈥� The practice saves educators time so they can 鈥渕ore easily differentiate their re-teaching or follow-up, especially when addressing common errors across the class.鈥�

Robin Lake, director of the Center on Reinventing Public Education, recently looked at the evidence base for using AI in math instruction, including whether it can help the “shocking number of students” with foundational skills’ gaps like those identified in a recent study. 

The May 13 analysis by TNTP found that almost half of the students sampled started the class with only one-third of the concepts and skills needed from earlier grades. Lake said AI can be used by schools to identify children who are struggling 鈥� and, at least to some degree, by the students themselves. 

鈥淎I can be very helpful in analyzing data and identifying gaps in student learning,鈥� she said.

And, if a student wants to learn a mathematical concept in a different way than what they鈥檝e experienced in class, she said, AI can provide a valuable alternative. 

鈥淎 lot of students are already doing this,鈥� Lake said. 

to use the technology, though many educators Terrie Galanti, associate professor at the University of North Florida, said AI success in student learning depends on how teachers are prepared to use it. 

鈥淎I can be more than an explainer or an answer giver,鈥� said Galanti, who teaches secondary mathematics and STEM integration/computational thinking. 鈥淲ith thoughtful prompts, AI can become part of interactive, collaborative conversations to deepen mathematics understanding.鈥� 

鈥嬧�婽he National Council of Teachers of Mathematics said in that teachers have long been accustomed to technological advances that change the way students learn. 

They had already adjusted to the availability of pocket calculators in the early 鈥�80s and, more recently, to the widespread use of , a mobile app that recognizes and solves math problems. 

It notes that advancements in AI make teachers more, not less valuable, in student learning. 

Latrenda Knighten, the organization鈥檚 president, told in March that students will still need to rely upon their own discernment to solve mathematical problems 鈥� regardless of what tools become available.  

鈥淲e know that children learn math from being able to problem-solve, being able to use reasoning skills, critical thinking, having opportunities to collaborate with each other and talk about what they鈥檙e doing,鈥� Knighten said. 

Irina Lyublinskaya, professor in the department of mathematics, science, and technology at Teachers College, Columbia University, distinguished between chatbots like ChatGPT and computational knowledge engines like She noted math specific AI-powered applications 鈥� including WolframAlpha and Symbolab 鈥� work very well. 

鈥淎I chatbots can help students learn math, and they can help teachers to support students, but this is not about asking ChatGPT to solve a math problem,鈥� she said. 鈥淚 know of research-based initiatives that use AI to adapt learning materials to students’ learning styles and abilities and these definitely help students learn.鈥�

One, she noted, was , developed by researchers and educators in Europe, and is now being tested in NYC. 

鈥淐hatbots can be trained as teaching assistants or tutors that can provide students proper scaffolding and feedback, helping them to learn math the same way they would with a real person,鈥� she said. 

Zachary A. Pardos is an associate professor of education at the University of California Berkeley where he studies adaptive learning and AI. He found, in conducted a year ago, that 25% of the answers provided by ChatGPT in algebra were incorrect. 

鈥淭hat鈥檚 pretty high,鈥� he noted. 鈥淢uch higher than you would want.鈥�

But the technology has improved since then. 

鈥淲ith the right techniques 鈥� at least in algebra 鈥� from an error perspective, I feel it is ready for real-time intervention in math,鈥� he said.

Inserting brief quiz questions into an online lecture can boost learning and may reduce racial achievement gaps, even when students are tuning in remotely in a distracting environment.

Through a unique collaboration between the and , the community now ranks 12th in state literacy, up from 31st just four years ago. This dramatic improvement wasn’t magic 鈥� it was the result of a systematic, community-wide approach to literacy that could serve as a model for districts nationwide.

The key to our success? A comprehensive public-private partnership that treats literacy as a community mission rather than just a school district initiative. The Learning Alliance, a nonprofit based in Vero Beach, has created an integrated support system that extends from birth through elementary school and beyond.

Its partnership with the school district works because it addresses literacy from multiple angles all in service of one goal 鈥� 90% of students reading on grade level by the end of third grade. 

To achieve this goal, the alliance funds 25 literacy coaches and reading interventionist positions within the district, ensuring that every school has dedicated expertise to implement the science of reading in grades K through 3. The district provides similar support in older grades. 

But that鈥檚 not all. The nonprofit engages families with children from birth to age 5, providing more than 5,500 age-appropriate books and 1,700 learning kits annually to under-resourced families. It also supports robust after-school and summer tutoring programs through the Moonshot Academy, where students show 50% more growth in reading compared to their peers not in the program. The afterschool program runs in the district schools, largely with district teachers compensated for their extra work by the Alliance.

The results speak for themselves. Indian River has doubled the percentage of A- and B-rated schools from 47% to 95%, and our third-grade reading scores now outperform the state average by nine percentage points (64% versus 55%). These aren鈥檛 just statistics 鈥� they represent thousands of children who now have the foundational skills they need to succeed in school and life.

Critical to this success has been the Moonshot Community Action Network, a coalition of over 150 local leaders who ensure that early literacy remains a community priority. This network includes business leaders, healthcare providers, faith-based organizations, and community advocates who understand that literacy is fundamental to our community鈥檚 future prosperity.

For superintendents and district leaders reading this, we offer several practical recommendations:

• First, look beyond traditional funding models. While public education funding is essential, strategic partnerships with community organizations can provide both financial resources and expertise that complement district capabilities.

• Second, invest in literacy coaches and reading interventionists. Having dedicated literacy experts in every school creates a support system for teachers and ensures consistent implementation of evidence-based reading instruction.

• Third, extend your teaching time. Our Moonshot Academy afterschool program creates opportunities for students to make more progress in less time. It pairs intensive tutoring with enrichment activities to boost engagement, and it works: students in the afterschool program average at least 50% more growth in reading than peers who do not participate.听

• Fourth, expand your reach beyond school walls. The family partnerships program demonstrates that literacy support must begin before kindergarten and continue outside school hours to be truly effective.

• Fifth, build community coalitions. The broader community鈥檚 investment in literacy creates a sustainable ecosystem of support that survives changes in school leadership or funding fluctuations.

For philanthropists and community organizations, think beyond traditional grant-making. The most effective partnerships involve deep collaboration with schools, shared accountability for outcomes, and a long-term commitment to the community.

Our journey hasn鈥檛 been without challenges, but it鈥檚 proven that significant improvements in literacy are possible with the right partnership model. The students鈥� success isn鈥檛 just about test scores 鈥� it鈥檚 about creating a foundation for lifelong learning and opportunity.

The pandemic may have created unprecedented challenges for education, but it has also shown us the power of community collaboration. In Indian River County, we鈥檝e demonstrated that when schools and community organizations work together with shared purpose and accountability, we can achieve remarkable results.

The question isn鈥檛 whether this model can work. The question is: Who will be next to replicate it?

In March 2020, resilience, innovation and adaptability became urgent priorities, backed by billions in federal funding. It was a Sputnik moment for American education.

We blew it.

We failed to take advantage of the moment. Instead of embracing lasting change, most school systems rushed back to “normal” 鈥� as if normal had ever been good enough.

The results are horrifying. Student achievement is in free fall. Fewer than one-third of students scored proficient in reading and math, according to the latest National Assessment of Educational Progress. These declines predated the pandemic but were exacerbated by prolonged school closures.

Given these realities, can policymakers still pretend the traditional education model works? A system designed over a century ago to train students for farm and factory labor is woefully inadequate for today鈥檚 needs. It cannot deliver the personalized learning students require in the Age of Artificial Intelligence.

This outdated system relies on one-size-fits-all solutions while assuming teachers can somehow provide differentiated support for every student. It rests on an increasingly fragile social contract: that students will attend school daily, that marginalized families will trust and wait for better service,and that schools are the sole places for learning. The pandemic shattered these assumptions.

The U.S. must rethink education. On this, the fifth anniversary of the start of the pandemic, the Center on Reinventing Public Education has launched , a forum for exploring bold, new ideas. Phoenix Rising looks back on the root causes of the disastrous pandemic response and articulates a vision for a more nimble, personalized, joyful and evidence-based public education system. Five years after the pandemic began, we reflect on the failures and propose a path forward.

Our research identifies key failures in the pandemic response and recovery:

• Schools lacked incentives, autonomy and capacity to deliver the personalized instruction needed to accelerate learning.
  
• States and the federal government provided little leadership, leaving districts to fend for themselves.
  
• Politics, not science, dictated too many decisions.
  
• Federal aid was distributed without clear expectations or accountability, offering only temporary relief.

The consequences are clear: declining test scores, wildly varied student needs within classrooms, disruptive behavior, chronic absenteeism and increasing mental health challenges for both students and teachers. Parents remain unaware of the full extent of learning loss, and public trust in education is eroding.

Rather than blame educators or school districts, we at CRPE diagnose a deeper problem: The education delivery system is fundamentally overmatched by its challenges. It cannot deliver the outcomes today鈥檚 students need.

We propose a future-ready system that prioritizes:

• Providing flexible, personalized learning pathways: Schools should act as portfolio managers, offering students personalized learning options rather than delivering all the instruction and support themselves. Core academics would remain in assigned schools, but students could use public dollars for apprenticeships, enrichment programs, tutoring and mental health support.
  
• Breaking down barriers in schools: Schools must dismantle rigid structures that limit student potential. Advanced coursework should be more accessible. Universal design for learning and individualized pathways to college and careers should be the norm, not the exception.
  
• Preparing students for the future: Success after high school requires more than career pathways, internships or college applications. Schools must emphasize durable skills like critical thinking, communication and leadership. By high school, students should be immersed in career exploration and have universal access to early college.
  
• Rethinking teacher roles and instruction: New schooling models should encourage team-based teaching. Evidence-based instructional practices must become standard. Research-based methods for reading, writing, math and behavior regulation should be integrated into teacher preparation and school support structures.

Forty years ago, CRPE advocated for a portfolio system of governance, where school boards diversified their offerings 鈥� traditional public schools, magnets and charters 鈥� while focusing on core services like funding and accountability.

Managing personalized pathways requires going further. It demands not just new governance structures, but also transformed instruction and student support.

States and localities must unlock funding, teacher assignments and student intervention strategies to enable innovative approaches. They should empower new governing bodies, whether independent boards, mayors or state-appointed leaders, to integrate ideas from outside the traditional district framework.

This transformation required bold action. Simply calling for more patience, more money and less regulation is not enough. Schools need sustained state leadership. With the federal government pulling back from education oversight, states must step up. Empty declarations of emergency won鈥檛 suffice. Top-down mandates won鈥檛 work.

Students can and will learn if given the chance 鈥� but only if educators rethink how they learn. That means transforming classroom instruction, teacher roles, technology use and more. States must reallocate federal funding flexibly, revamp laws to incentivize innovation and create new opportunities for experimentation beyond the traditional system.

Above all, the next wave of education reform must look forward, not backward. American schools cannot afford to cling to outdated structures out of a misguided allegiance to the past.

Policymakers must empower schools to embrace new ideas, act on evidence and be bolder in pursuing better outcomes.

Students鈥� futures 鈥� and the country’s economic and social prosperity 鈥� depend on it.

Perhaps you already knew that?

In a series of posts in 2020 and 2021, I wrote about the research on past educational disruptions and predicted what they might mean for children going through COVID-19.

This month marks the fifth anniversary of the pandemic. What have we learned since then? Here鈥檚 my analysis of what I got right and what I got wrong:

Prediction No. 1: Lost learning time will translate into lost learning.

As I wrote in 2021, 鈥� and sometimes even after a disruption in schooling, researchers are able to detect noticeable differences in student outcomes.鈥�

COVID’s effects were so large that they showed up immediately. And they were worse for students who missed more school. For example, a recent in Nature concluded that, 鈥渃ountries with the shortest closures experienced relatively small losses鈥� and 鈥渃ountries with the longest closures, experienced losses of 鈥� 9 to 12 months of learning.鈥� Here in the States, a team of researchers led by University of Washington economist Dan Goldhaber that remote instruction was the 鈥減rimary driver of widening achievement gaps.鈥�

Predictions No. 2 and 3: The losses are likely to be large. Their full extent may not show up immediately, but small losses can grow over time.

In 2020 and 2021, I was looking at localized events like an earthquake in Pakistan or teacher strikes in Argentina. Researchers found noticeable negative effects from these events, but there was no modern precedent for the scale and length of the COVID-related disruptions.

Here in the States, the average student by the equivalent of half a grade level in math and one-third of a grade level in reading. Today, with federal COVID funding ending, student performance is still far below where it was.

Prediction No. 4: Learning losses are likely to be larger in younger children.

Based on prior research, I anticipated that kids in middle school and high school might transition to remote learning better than their younger siblings. But it鈥檚 hard to know which age group ultimately struggled the most. scores for high schoolers fell to 30-year lows, but so did fourth and eighth grade math and reading scores. Curriculum Associates data point to ongoing gaps among younger kids, while NWEA data middle schoolers are the furthest behind. In other words, there鈥檚 no biggest loser here; achievement is down across the board.  

Prediction No. 5: Math scores are likely to drop the most.

Math skills are generally picked up at school, while in English Language Arts, proficiency is more closely linked to a student鈥檚 home environment. That shows up in the data: Math scores declined more when students were learning remotely and have shown more signs of recovery since then.

Meanwhile, reading scores have continued to decline. That could partly reflect broader societal trends, but state leaders should be looking toward Mississippi and other Southern states for ideas on how to get those back on track.

Prediction No. 6: Beyond academic losses, students are at risk of disconnecting from education.

In 1916, schools in many parts of the country closed for weeks in the midst of a polio outbreak. Researchers later that those shutdowns caused some students to drop out of school and never return.

More than a century later, COVID magnified this to an extreme, changing the relationship kids had with school and leading to in chronic absenteeism. While those rates have come down a bit from their highs in 2022, I didn鈥檛 anticipate just how much COVID would break cultural norms around school attendance, and how hard it would be for schools to restore those (good) habits.

Prediction No, 7: Higher-income students may not suffer any noticeable effects.

I underappreciated how much a district鈥檚 decisions about in-person versus remote schooling would have on all students. The Goldhaber found that, 鈥渆ven at low-poverty (high income) schools, students fell behind growth expectations when their schools went remote or hybrid.鈥�

Still, higher-income and higher-performing students have managed the post-pandemic recovery better than others. For example, in eighth-grade math, the top 10% of students made noticeable gains from 2022 to 2024, while scores continued to fall for lower-performing students. At the high school level, the percentage of students who took and passed an Advanced Placement test dipped in the wake of COVID but has pre-pandemic levels. Similarly, the of high school students who are also taking credit-bearing college courses is hitting all-time highs. In other words, even as average achievement scores are down, many more high school students are finding ways to take more advanced courses.

However, this leads to 鈥�

Prediction No. 8: Low-income and disadvantaged students will suffer the biggest losses.

The virus may have been the same, but it did not affect everyone equally. As I last year, all kids missed more school in the wake of the pandemic, but those increases varied substantially. For example, the kids with the best attendance records missed about one extra day of school per year, while the kids with the worst absentee rates missed multiple weeks worth of school time.

The same trends appear in achievement scores: The bottom has fallen out across a variety of  tests, grades and subjects.

Prediction No 9: The COVID-induced recession will affect children, families and schools in many ways.

The pandemic鈥檚 effects on students have the potential to be long-lasting. For example, the of teacher strikes in Argentina found that the children of strike-affected students were more likely than their peers to be held back in school. That is, the effects passed on to later generations.

But beyond the students themselves, no one could have anticipated all the downstream effects of the COVID school closures. They are at least part of the story behind public school enrollment declines, families moving  away from and , and declining satisfaction with the nation鈥檚 public schools, not to mention political realignments.

Prediction No. 10: Without any action, the losses are likely to have long-term consequences.

To its credit, Congress provided states and school districts with $190 billion in 2020 and 2021. As hard as it is to fathom, the best research suggests it would take a lot more money to get kids fully back on track. The achievement declines were that large.

The money is now gone, and the national conversation has moved on. But I鈥檓 struck by what I wrote a few years ago: that if policymakers don鈥檛 act to get kids back on track, they will be, 鈥渃ondemning a generation of children to worse academic and economic outcomes throughout their life.鈥�

FutureEd analyzed speeches from 41 governors to identify states’ education agendas for the coming year, highlighting common themes, bipartisan commitments and partisan divides. Across party lines, governors remained committed to investing in public education, with many proposing increased K-12 funding and efforts to modernize school finance formulas to better support high-need students. Alongside these general financial commitments, governors prioritized strengthening the teaching profession, addressing youth mental health, restricting cellphone use and expanding career pathways for high schoolers. 

Still, few governors proposed new steps to directly improve student learning. Some emphasized efforts to raise academic standards and strengthen accountability, topics absent in recent years. But this year鈥檚 speeches largely sidelined new curriculum initiatives, perhaps because many states have enacted reforms in that area 鈥� particularly in literacy 鈥� in recent years. Mentions of academic acceleration programs that were widely supported during the pandemic but now face an uncertain future as federal ESSER funds expire were also rare. These include high-dosage tutoring, afterschool and summer enrichment. Some governors argued that their school choice initiatives would improve student outcomes. 

School choice remained a key point of division, with several Republican governors advocating for more private options that Democrats opposed. Ideological divides also resurfaced on race and gender 鈥� topics largely absent from speeches in recent years 鈥� as a few Republicans called for banning diversity, equity and inclusion (DEI) initiatives, restricting transgender participation in sports and opposing what they described as 鈥渋ndoctrination鈥� and 鈥渨oke agendas鈥� in education.

Here is a summary of the major education proposals in the governors’ addresses (click on each state in the interactive maps below): 

Student Well-Being

One of the most popular topics this year was student well-being, with a focus on mental health, technology use, school meals and safety. While youth mental health remains a top concern 鈥� highlighted by 15 governors 鈥� many are shifting from traditional investments in counseling and school-based services to restrictions on cellphones and social media.

Concerns over technology鈥檚 impact on student mental health are growing. As Arkansas Gov. Sarah Huckabee Sanders noted in her annual address, 鈥淪uicide rates among teens have tripled, self-harm among girls has risen by nearly 200%, and depression among teenagers has increased by 150%. The culprit is clear: unrestricted access to phones and social media.鈥� Huckabee Sanders and the leaders of at least 12 other states, mostly Republican-led, proposed bans on phones in schools, with most citing mental and behavioral health concerns. Governors also pushed for stronger internet safety measures and social media restrictions.

In New York, Democrat Kathy Hochul has taken a different approach with her “Unplug and Play” initiative, which expands outdoor and extracurricular opportunities to reduce children’s reliance on social media. She also called for shielding students from the risks posed by artificial intelligence. 

While cell phone and social media restrictions have gained some bipartisan traction, major investments in school-based mental health are largely a Democratic priority this year. Wisconsin Gov. Tony Evers, for example, proposed nearly $300 million to expand mental health services, including suicide-prevention programs and educator training.

To address students鈥� nutritional needs, Democratic governors in Maine, Kansas, New York and Wisconsin proposed expanding free school meals. On the Republican side, Huckabee Sanders introduced the Buy Local Act to encourage schools to purchase their food in state and proposed using medical marijuana revenue for meal funding, and North Dakota Gov. Kelly Armstrong pledged to expand his state鈥檚 Farm-to-School program. 

Governors from six states 鈥� Indiana, Georgia, Texas, Missouri, New Mexico and South Carolina 鈥� proposed school safety efforts. Indiana Gov. Mike Braun plans to establish an Office of School Safety, Gov. Henry McMaster wants to fund school resource officers in every South Carolina school and Missouri Gov. Mike Kehoe is pursuing legislation requiring regular safety planning meetings and incident reporting.

School Choice

School choice remains a key topic this year, with 15 governors addressing the issue. While initiatives to let families use public money for private schooling dominated the discussion, several governors proposed expanding public-school choice, sometimes alongside private-school initiatives.

Proposals largely followed party lines. Nine Republican governors advocated for expanding education savings accounts and other private school-choice initiatives. McMaster aims to pass new ESA legislation and allocate $30 million after the South Carolina Supreme Court ruled the state’s previous program unconstitutional.

Braun seeks to make private school choice available to every student in Indiana, while Kehoe proposes adding $50 million in state funding to supplement a tax credit-funded ESA program. Virginia鈥檚 Glenn Youngkin is requesting $50 million for 鈥淥pportunity Scholarships鈥� for low-income families, and Texas Gov. Greg Abbott is renewing his push for a long-debated universal school choice measure. 

In the wake of widespread criticism of Arizona鈥檚 costly education savings account program and its lack of transparency on spending and student performance, some Republicans emphasized stronger oversight and protections for public school funding in their proposals. Idaho鈥檚 Brad Little stressed the need for a 鈥渇air, responsible, transparent and accountable鈥� approach that prioritizes families in need without diverting funds from public schools. Governors in South Dakota and Virginia also emphasized safeguarding public school resources.

Meanwhile, three Democratic governors 鈥� Arizona鈥檚 Katie Hobbs, Kansas鈥檚 Laura Kelly and Kentucky鈥檚 Andy Beshear 鈥� expressed opposition to using public funds for private education. Hobbs is once again seeking to add guardrails to Arizona鈥檚 universal ESA program, which was enacted by her Republican predecessor. These include setting income caps and ensuring transparency in spending. Kelly and Beshear vowed to block new private school choice initiatives.

At the same time, six governors from both parties are pushing to expand public-school choice. Colorado鈥檚 Democratic Gov. Jared Polis voiced support for giving parents more options and making it easier for high-quality and innovative schools to open.

Republicans Joe Lombardo of Nevada and Kehoe supported allowing students to attend public schools outside their assigned attendance zone. Armstrong proposed a new policy that would allow charters to open in North Dakota.

Higher Education

Fifteen governors outlined higher education priorities focused on affordability, workforce alignment and alternative pathways. McMaster pledged to freeze in-state tuition for a sixth year and is seeking $80 million in financial aid to ensure all eligible students receive state assistance. Hochul proposed free community college for New York students entering high-demand fields like advanced manufacturing, education and health care.

Several states are expanding access through alternative postsecondary pathways. Huckabee Sanders launched the ACCESS initiative, which, among other things, will expand scholarships to include associate degrees and non-degree programs, and fund college credits for high school students. Rhode Island Gov. Dan McKee plans to fund dual enrollment, allowing high school students to earn college credit. Connecticut Gov. Ned Lamont is similarly focused on adding more dual-enrollment and 鈥渟tackable鈥� credentials. 

Huckabee Sanders and Abbott proposed banning DEI programs and preventing 鈥渋ndoctrination鈥� from professors. 鈥淲e must purge it from every corner of our schools and return the focus to merit,鈥� said Abbott of the state鈥檚 public universities.

Early Education and Child Care

Thirteen governors highlighted early childhood education and care, focusing on expanding access, improving affordability and addressing workforce challenges.

Democratic governors in New York, New Mexico and Kentucky promoted universal early childhood programs. Arizona Democrat Hobbs was among five state leaders from both parties who prioritized affordability, proposing the Working Families Child Care Act to lower the cost of care by two-thirds. Kehoe, a Republican, proposed a $10 million grant program to foster partnerships among Missouri employers, community organizations and child-care providers to expand access through collaborative solutions.

In Iowa, Republican Gov. Kim Reynolds introduced a continuum-of-care initiative to integrate preschool and child-care services, offering grants to providers that coordinate resources like shared staffing and transportation.

Several state leaders also addressed child care workforce shortages. Gov. Bill Lee proposed expanding eligibility for Tennessee’s WAGES program that provides early childhood educators with annual salary supplements of up to $7,800 based on their education level, while Reynolds launched a statewide fund to encourage donations from individuals and businesses to support child care worker wages.

Workforce Development and Career Education

Twelve governors highlighted workforce development and career education, focusing on expanding high school career training, strengthening industry partnerships and aligning postsecondary programs with workforce needs. More than half of these efforts target K-12 students.

Several states are prioritizing career training in high schools. In Indiana, Braun plans to expand partnerships between employers and high schools, while Connecticut鈥檚 Lamont is expanding the state鈥檚 youth service corps, offering students internships or paid apprenticeships at local businesses. Lee seeks to double participation in Tennessee鈥檚 Youth Employment Program, making it year-round to provide meaningful work experience, particularly for underserved communities. Kehoe wants a $1 million investment to expand high school career counseling and launched the Governor鈥檚 Workforce of the Future Challenge to improve coordination among K-12 schools, businesses and colleges. Youngkin is expanding 鈥渓ab school鈥� partnerships with historically Black colleges and universities specializing in health care, coding, space, maritime industries and teaching.

Other governors are focusing on postsecondary workforce training. McMaster is requesting $95 million in lottery funds for Workforce Industry Needs Scholarships through South Carolina’s technical college system. Lee has proposed the Tennessee Works Scholarship, which would cover tuition, fees and essential resources for students at the Tennessee College of Applied Technology. And McKee introduced Ready to Build, a career and technical education program at the Community College of Rhode Island, designed to create a pre-apprenticeship pathway into building trades jobs.

The Teaching Profession

A dozen governors discussed initiatives to improve the teaching profession, with most emphasizing higher teacher pay. Eight 鈥� primarily Republicans 鈥� proposed salary increases. McMaster accelerated the state鈥檚 plan to set a $50,000 minimum salary, moving the target from 2026 to this year. Braun aims to raise the minimum salary in Indiana by $5,000, bringing it to $45,000. 

At least two states are prioritizing pay increases for top-performing teachers. Abbott seeks to boost average teacher pay to a 鈥渞ecord high鈥� while expanding the state鈥檚 Teacher Incentive Allotment program, which allows top educators to earn six-figure salaries. Similarly, Nevada鈥檚 Lombardo plans to reward high-performing teachers through the Excellence in Education Fund under the Nevada Accountability in Education Act. Beshear was the only Democrat to mention teacher pay.

Democrats Lamont and Wes Moore of Maryland proposed expanding affordable, debt-free pathways into teaching through apprenticeships and other means. 

Student Learning 

Only 10 governors discussed curriculum and instruction initiatives for the upcoming year. Among them, Reynolds proposed one of the few major policy pushes: a comprehensive math bill aimed at identifying struggling students, providing personalized support and strengthening instruction through evidence-based professional development and high-quality teacher preparation.

Lombardo proposed the Nevada Accountability in Education Act, a comprehensive initiative that would, among other things, demand 鈥渟tricter accountability, equity and excellence鈥�, focus on literacy and direct resources to struggling schools. He didn鈥檛 detail how he planned to increase accountability for student results. 

Nevada is one of seven states, along with Oregon, Idaho, Mississippi, Missouri, Michigan and Massachusetts, prioritizing standards and accountability this year. Oregon Democratic Gov. Tina Kotek committed to making school and district accountability a central focus of her 2025 agenda, proposing to increase transparency through a statewide, publicly accessible student information system, while Mississippi Republican Tate Reeves proposed raising academic standards and overhauling the state鈥檚 school grading system. Michigan Gov. Gretchen Whitmer, a Democrat, has proposed the Students, Metrics and Results with Transparency (SMART) plan, which would direct investments toward underperforming schools and improve transparency to better inform parents about student performance.

In Massachusetts, which recently eliminated its graduation assessment requirement, Democrat Maura Healey called for a Statewide Graduation Requirement Council to establish a new high standard. 鈥淪tudents, families and employers need to know what a diploma represents,鈥� she stated. 

Additionally, three Democratic governors emphasized investments in expanded learning time. McKee proposed $2.5 million in grants for out-of-school learning in Rhode Island, Kotek aims to continue funding summer programs and New Mexico Gov. Michelle Lujan Grisham wants to expand summer literacy initiatives. 

Meghan Gallagher of 社区黑料 developed the interactive maps. FutureEd Research Associate Tony Tao contributed to this analysis. 

Fortunately, federal funding structures include a little-used tool for sustaining student support programs. , which provides billions in funding for schools serving low-income students, includes allowing states to set aside up to 3 percent of their allocations for direct student services, including and other high-impact interventions. This funding could be a lifeline to the all-too-recently defunded programs making the on the most . 

This provision requires no new funding yet remains largely untapped. Policymakers and education leaders have long pointed to the 3% state set-aside in Title I as a way to sustain critical interventions, but few states have taken full advantage of it. Instead, they simply add it to the money they share with districts.

But with pandemic-era funds largely depleted and budgets tightening, this overlooked mechanism offers a rare opportunity: preserving the programs that have made the greatest impact without forcing districts into painful trade-offs.

For now, Title I appears to be a stable source of federal funding, with regular increases each year. Even when the Trump administration moved to halt federal funding last month, Title I stayed online. Since 1980, allocations to districts have grown by an average of 4.2% a year. This steady rise creates a practical opportunity: A state could allocate its full 3 percent set-aside for direct student services while still increasing district funding, at least in nominal dollars.

For example, if all states had reserved the complete 3% entitled in 2023, they would have had $552 million to continue supporting direct student services and still raised allocations to districts by $639 million. And 2023 allocations weren鈥檛 an anomaly. As Figure 1 shows, growth exceeds 3% in most years, with 24 of the past 43 years surpassing this mark and 11 others showing growth below 3%.

However, even looking at the worst years for Title I growth, it is still feasible to build toward the 3% allocation over a series of years. A state could gradually build up to that level, taking 1% initially and then additional shares in two subsequent years. The transition would be helped by the fact that a districts’ Title I allocations fluctuate from year to year reflecting changes in the U.S. Census Bureau estimates of the number of eligible children in each district.

Ohio is already leveraging the set-aside allocation with its grant. The program directs funding toward high-dosage tutoring, advanced coursework, career pathways, personalized learning, and academic acceleration, aligning with local improvement plans to improve student outcomes. 

Districts have leveraged this grant funding to introduce Advanced Placement and College Credit Plus courses, expanding access to high-level curriculum that were previously out of reach. At the same time, the grant has helped establish career pathways that expose students鈥攕ome as early as middle school鈥攖o high-demand industries, giving them a head start on college and career readiness. 

These investments not only accelerate learning but also work to close long-standing opportunity gaps, illustrating the potential of targeted Title I set-aside spending to drive meaningful change.

Beyond successful state examples, the Council of Chief State School Officers has published a addressing the logistics of implementing the set-aside allocations. The guide also details how, with proper planning and careful communication, states can use this money as a powerful lever for filling gaps in critical supports, like intensive tutoring and wraparound services. Done right, this is more than just financial maneuvering; it鈥檚 a blueprint for how federal education policy can be both ambitious and effective.

The expiration of the $190 billion in federal Elementary and Secondary School Emergency Relief (ESSER) funding marks a natural transition, but it need not mean the end of direct student services, such as tutoring programs, that have proven valuable to mitigating pandemic-era learning loss. 

By leveraging year-over-year increases in Title I funding, states can establish a sustainable mechanism to continue these services without reducing district budgets. This approach balances our commitment to students with our responsibility to districts, ensuring that we move forward in a way that is both effective and equitable.

The end of ESSER doesn鈥檛 have to mean the end of targeted academic support. But making the most of existing resources requires political will. The question is whether states will act before students fall even further behind.

One of the most striking features of the troubling results from the most recent National Assessment of Educational Progress is the much lower percentage of students scoring proficient on NAEP than on many states鈥� own 2024 standardized exams.

By now, you鈥檝e likely seen the results: modest improvements in math, but not enough to get students back to pre-pandemic performance levels; fourth graders fell further behind in reading; a record 34% of eighth graders scored 鈥渂elow basic鈥� in reading. 

In addition to the national summaries, NAEP reported student achievement in each state, where proficiency rates ranged from a high of 51% in fourth-grade math in Massachusetts to a low of 14% in eighth-grade math in New Mexico.

States, of course, are required by federal law to administer their own annual standardized tests in math and reading. FutureEd students鈥� performance on NAEP and on their states鈥� tests and found that, for the most part, students met proficiency standards at significantly higher rates on their states鈥� exams, especially in reading.

The gaps were at least 15 percentage points in three-quarters of the states. In some, they were even greater. Seventy-two percent of Virginia鈥檚 eighth graders were proficient in reading, more than double the percentage on NAEP. Iowa reported more than three-fourths of its eighth graders proficient in reading in 2024, compared with less than a third of the state鈥檚 students on NAEP. We also found that the gaps increased in many states between 2022 and 2024, including in 26 states in fourth-grade reading and 22 states in eighth-grade reading.  

Why is there so much misalignment between NAEP and state results?

Perhaps more than any other factor, it鈥檚 lower state standards.

To achieve proficiency on the national assessment, students must show 鈥渟olid academic performance and competency over challenging subject matter.鈥� That鈥檚 where Rhode Island, Massachusetts and the District of Columbia set their proficiency bar. But most states’ of that benchmark, landing within the range of NAEP鈥檚 lower “basic” standard, which requires students to demonstrate only 鈥減artial mastery of fundamental knowledge and skills.鈥� In Virginia 鈥� which has introduced new academic standards 鈥� and Iowa, the bar for reading falls below even that.

What鈥檚 more, in  and other states, students can be performing 鈥渙n grade level鈥� without meeting the state鈥檚 鈥減roficient鈥� standard in the subject they鈥檙e studying. And some states have gone further, lowering the passing grades on some or all of their standardized tests in recent years.

The Oklahoma State Department of Education reported significant gains in 2024, including a 24-point jump in the percentage of students achieving proficiency in fourth-grade reading since 2022 and across-the-board improvements over pre-pandemic levels. But the gains coincided with a lowering of the state’s proficiency standards, which officials didn鈥檛 publicize when they released the improved test scores.  obtained by an Oklahoma news organization revealed that the 2024 scores would have been the same as or slightly lower than 2023 results if the standards had remained the same. On NAEP, Oklahoma鈥檚 proficiency rates declined in reading and improved slightly in math between 2022 and 2024, but they remained below pre-pandemic levels.

Similarly, New York reported across-the-board improvements in student achievement in 2024 after  in 2023. But these gains were not mirrored on all of the state鈥檚 2024 NAEP results. Wisconsin also registered higher proficiency rates on its 2024 assessments after , only to have most of its NAEP scores decline in 2024. This points to the value of an independent national measure of student achievement, like NAEP.

One of the more troubling findings from the 2024 state assessment cycle is the wide gap in proficiency rates between fourth and eighth grades, with eighth graders, on average, performing much worse than their younger counterparts. The gaps are far more pronounced in math than in reading. In New Jersey, for example, 45% of fourth graders were proficient, compared with only 19% of eighth graders. Similarly, in Washington, D.C., 29% of fourth graders and just 12% of eighth graders achieved proficiency.

With many schools struggling to return to performance levels that were declining even before COVID’s disruptions, having an accurate measure of achievement is critical. Aligning more state proficiency standards with NAEP’s would increase transparency and make it easier for everyone 鈥� students, parents, teachers, administrators and elected officials 鈥� to be clear on where every state needs to focus to improve educational outcomes for all students. 

Released Tuesday morning, is the latest dispatch from the , a data project led by a team of researchers at Dartmouth, Harvard, Stanford, and the testing group NWEA. In two studies released last year, the consortium unearthed in high-poverty areas since 2020, along with resulting from billions of dollars in federal assistance to K鈥�12 schools. 

This week鈥檚 update comes on the heels of a disheartening publication of test scores from the National Assessment of Educational Progress, often referred to as the Nation鈥檚 Report Card. While some had hoped that results from that exam would provide reason for hope, only minimal progress was made in fourth-grade math; reading scores were actually worse than in 2022, the nadir of the pandemic. 

Thomas Kane, a professor of economics and education at Harvard, compared the sustained learning loss of the last few years with 鈥渢he tsunami following the earthquake鈥� 鈥� a destructive after-effect that has almost entirely resisted remediation efforts by local, state, and federal authorities. Struggling students, in particular, have fallen further behind their higher-performing peers, he observed.

鈥淕iven all the money that’s been spent, and the fact that students already lost ground between 2019 and 2022, you would have expected that there would be some bounce-back in reading,鈥� Kane said. 鈥淏ut no, actually. Students continued to lose ground, especially at the bottom end.鈥� 

While NAEP offers state-by-state comparisons, along with the results from several dozen major urban districts, the Scorecard group combines those figures with local testing data for 35 million students across 43 states, allowing the public to chart the trajectories of individual districts since 2019. 

Across the country, Kane and his collaborators calculate, just 11 percent of students in grades 3鈥�8 are currently enrolled in districts where average reading levels exceed those measured in 2019; 17 percent are in districts where math knowledge is higher than the last pre-pandemic year. Set against the continuing fall in literacy, a slight rebound in math scores 鈥� about one-tenth of one grade level since 2022 鈥� represents most of the good news. 

In relatively poorer communities, that silver lining is almost entirely accounted for by federal ESSER funds, which totaled $190 billion between 2021 and 2024. The report indicates that those grants prevented an even greater freefall in learning, while noting that 鈥渢here were higher-impact ways to use the dollars鈥� to speed student recovery.

Rebecca Sibilia is the founder of , a research and advocacy group that advocates for more and better-designed resources for schools. A frequent critic of the quality of school finance data, she said the breakneck pace at which ESSER dollars were appropriated and distributed made it virtually impossible for them to be maximally effective.

“We absolutely have research that shows money matters, and helps us understand how money matters,鈥� she said. 鈥淓SSER was not constructed in a way that aligns with that research.鈥�

Michael Petrilli, president of the right-leaning Thomas B. Fordham Institute, called the Scorecard study 鈥渄evastating.

鈥淲e already knew that the bottom had fallen out for most states, but now we see how hard it is to find districts bucking the terrible trends,鈥� he wrote in an email.

鈥楾wo kinds of bad news鈥�

Perhaps the most alarming trend of the period bridging the COVID depths of 2022 and the present day has been a substantial rise in educational inequality. 

By sorting thousands of school districts according to their number of students eligible for free or reduced-price lunch (a commonly used proxy for poverty), the Scorecard researchers found that academic recovery over the last two years has proceeded much more quickly in affluent areas.

In nearly one-third of all low-poverty school districts, math performance has been restored to the pre-pandemic status quo; the same is true in just 8 percent of high-poverty districts. In all, over 14 percent of the richest districts (i.e., those where household income is higher than in 90 percent of other places) have returned to 2019-era learning in both math and reading, compared with less than 4 percent of the poorest districts. 

A similar dynamic has been apparent in NAEP scores going back more than a decade. While the 2010s saw gradually declining results on average, the highest-scoring students tended to make some progress in each administration of the exam. Meanwhile, their struggling classmates experienced much larger reversals. Since 2013, the disparity in fourth-grade reading performance between kids at the 90th and 10th percentiles, respectively, grew by 14 points; the divergence in eighth-grade math grew by 16 points over that decade.

Stanford sociologist Sean Reardon, who leads the Scorecard project alongside Kane, said the widening gaps make it clear that the task of general academic recovery must be accompanied by a special focus on students who are at risk of never getting back on track. 

“There’s two kinds of bad news between the NAEP results and ours,鈥� Reardon said. 鈥淥ne is the disappointing lack of recovery, and even continued decline, in reading. Those average trends are disappointing, but they’re compounded by the fact that the negative trends are worse for the kids in the highest-poverty districts.鈥�

The worrying class bifurcation is apparent from coast to coast, but Kane specifically identified achievement gaps in his home state of Massachusetts. There, the well-to-do Boston suburbs of Lexington and Newton have either surpassed their academic performance of a half-decade ago or have very nearly dug themselves out of the hole. 

Just a few miles away, however, in the working-class cities of Everett and Revere, the average student is floundering more than a year behind the pace set by similarly aged students just five years ago. In Lynn, one of the most troubled school districts in the state, elementary and middle schoolers are two years behind in math and over 1.5 years behind in reading.

The report includes from relatively disadvantaged communities (including Union City, New Jersey, Montgomery, Alabama, and Natchitoches Parish, Louisiana) that had made significant strides back to normalcy. But the typical such district still faces years of work to regain what was lost. 

Joshua Goodman, an economist at Boston University, said that education leaders needed to guard against the sense that emerging gaps simply represented the 鈥渘ew normal.鈥� If he鈥檇 been told in 2020 that children would still be scuffling to this extent by the middle of the decade, he said, he would have been shocked and disappointed. 

鈥淚 think I implicitly believed that, once the pandemic receded and schools reopened, the normal operation of kids’ lives would somehow cause them to bounce back,鈥� Goodman recalled. 鈥淚 don’t know if I was just being naive or not thinking it through properly, but this is a very grim result.”

Meager return from COVID funds

The dour note struck by observers is largely related to the meager returns of Washington鈥檚 relief efforts. 

Previous work from the Education Recovery Scorecard has pointed to a modest bump in student performance that followed an infusion of billions of dollars to states and districts. But that upward movement didn鈥檛 come close to reversing the full extent of COVID鈥檚 damage; for that, researchers estimated, hundreds of billions of dollars more would be needed.

With federal funds now expired, and no new federal appropriations on the horizon, ESSER鈥檚 final impact can begin to be measured. For every $1,000 spent per student between 2022 and 2024, the authors estimate, math scores increased by roughly .005 standard deviations (a scientific measure showing the distance from the statistical mean). 

In comparison with other policy changes in education, Kane and Reardon showed, this is a fairly small figure 鈥� just a tiny fraction of by schools that adopted the Success for All reform model, for example, or those that followed the implementation of high-dosage tutoring programs. 

Kane said the relatively freewheeling structure of ESSER funds 鈥� states were only required to spend 20 percent of the aid on programs specifically aimed at lifting student achievement 鈥� meant that many expenditures were not efficiently targeted at the schools and students of greatest need. The small payoff could serve as a warning to Republicans reportedly the Department of Education and disbursing its various revenue streams to states to spend freely. 

鈥淭his is an example of bypassing federal regulators, or even bypassing state regulators, and giving all the money directly to school districts,鈥� Kane argued. 鈥淲e just saw what happens: Some school districts will figure out how to use the money well, but others won’t.”

Referencing widely circulated papers by school finance researchers Kirabo Jackson and Eric Hanushek, Sibilia said the general case for spending more on K鈥�12 schools was sound. But ESSER money was sent out the door quickly, often to districts that didn鈥檛 serve large numbers of needy students. While spending it, district leaders had to make fast decisions with incomplete information.

The simultaneous and temporary explosion in districts鈥� budgets had led to a concurrent increase in shoddy vendors for services like tutoring and professional development. No matter the amount of money that Congress might have awarded, she added, the effects of ESSER would have been dampened by the limited supply of high-quality providers.

“There are a few researchers in the country that are dogmatic in saying that money, no matter how it’s spent, will give you a positive return,鈥� Sibilia said. 鈥淏ut I think 95 percent of the people studying money in education will tell you that spending is only as good as what you can buy.鈥�

Before fourth and eighth graders took the 2024 NAEP math assessment, they were asked how many days of school they had missed the previous month. Forty-nine percent of fourth-graders who would score at the 75^th percentile or higher on the test had missed no days the previous month, compared with 26% of those scoring below the 25^th percentile.

Equally striking, 45% of students in the bottom quartile reported missing three or more days of school in the previous month compared with just 20% of students in the top quartile. And at the extreme ends of the absenteeism spectrum, 7% of the lowest-performing eighth graders on the NAEP math test reported missing more than 10 days of school in the previous month, compared with just 1% of top scorers.

Correlation strongly suggests causation. It鈥檚 impossible to prove that students performed better because they were in school every day, but it鈥檚 the logical conclusion.

A detailed comparison of state test scores and student absenteeism by Rhode Island education officials suggests as much. They found that just 10% of students who had been chronically absent for three consecutive years scored proficient on Rhode Island鈥檚 own standardized math tests in 2024, and 13% were proficient in reading. In contrast, 40% of students who attended regularly were proficient in math and 38% were proficient in reading. As on the state’s dashboard, 鈥渓ong-term chronic absenteeism has a compounding negative impact on student performance.鈥�

Attendance influenced achievement significantly even among students facing the many challenges of poverty. While it’s hardly surprising that only 18% of Rhode Island鈥檚 low-income students who attended school regularly were proficient in reading and math last year, just 11% of those who were chronically absent were proficient in reading, and only 9% met that bar in math.

The upshot is there needs to be a relentless focus at the state and district levels 鈥� beyond the work of individual schools 鈥� on getting every student in school every day. Transparency is essential to progress. Rhode Island is the only state that publishes detailed, real-time attendance data for every one of its public schools, allowing officials to correlate state test scores and absenteeism.  More than a dozen states have yet to release attendance data from the 2023-24 school year, making it difficult for policymakers to even know which absenteeism problems they need to solve.

The quality of instructional materials, tutoring programs or new technology tools can’t make much of a difference if students aren鈥檛 in school.

I would argue that the causes are dual problems that are not only intertwined, but sit at the very core of this crisis of stalled learning recovery: Far too many boards of education have abandoned accountability and failed to embrace innovation. Yes, the challenges education stakeholders 鈥� parents, educators and policymakers alike 鈥� face are enormous, but the solutions are within reach if, collectively, we commit to a balanced approach. To recover lost ground, decisionmakers need to learn from bright spots in education, where accountability and innovation intersect to drive real results.

At the National Association of Charter School Authorizers we know high-performing charter schools can serve as models for driving meaningful results for students. Research shows how from authorizers contribute to charter school success and how high-performing charters have . NACSA鈥檚 own research has demonstrated the and approaches to learning that have found a home and are growing in the charter sector. Charter schools have consistently proven their ability to dramatically accelerate the learning of students furthest away from opportunity 鈥� a critical capacity, given the sobering NAEP outcomes for students academically behind. 

Here are three big lessons, based on the successes we have seen from authorizing high-quality charter schools, that decisionmakers can use to ensure all students receive an excellent education:

Understand accountability, encourage innovation and problem solving: For accountability to truly drive meaningful student results, there must be a shared understanding of what it is and is not. Accountability systems or frameworks ensure that all students meet high academic standards by defining expectations, creating conditions for innovation and accommodating the needs of families and communities to drive change when schools fall short. While accountability must impose consequences for failure 鈥� including making the tough decision to close schools that consistently fail to educate students well 鈥� it鈥檚 about so much more than that. It is about permitting excellent schools to expand so their successes can be replicated and, perhaps most importantly, ensuring that high and rigorous expectations for student learning are perpetuated.

When accountability is at its best, it fosters greater innovation. This is especially true in high-quality charter schools, where accountability and innovation are interwoven by design. These schools are given the flexibility to innovate and problem-solve in multiple ways 鈥� such as implementing evidence-based instructional models that draw from research-backed teaching strategies, finding creative ways of re-engaging students and using technology in unique ways 鈥� while being held to rigorous performance standards, ensuring they deliver for their students. It is this balance that needs to be more broadly replicated, allowing accountability and innovation to compliment one another in service of student learning and growth.

Align rigor with clear guidance: In order for teachers and school leaders 鈥� and, ultimately, students 鈥� to excel, they need policies and practices that promote achievement, which includes explicit guidance around expectations and clear measures of success.

Charter school authorizers play a critical role in ensuring expectations are both rigorous and attainable. By using tools like school performance frameworks and reports, progress monitoring systems, site visits, consistent parent communications and other evaluation and transparency systems, authorizers provide schools with the tools to understand how they鈥檙e performing and what鈥檚 expected of them. This alignment empowers educators to focus on what matters most: student learning.

Follow the excellence equation: High-performing charter schools have two important elements that drive their success: They are responsible for determining and executing their programmatic objectives, while their authorizers must set clear expectations and be transparent about performance. This clear delineation is the key to producing high-functioning schools where students are prepared for the next phase of learning and life.

At NACSA, we鈥檝e observed that this excellence equation 鈥� balancing accountability for outcomes with freedom to innovate 鈥� is a common feature of the highest-performing schools. All schools should be empowered to tailor their programs to meet their students’ needs while accountability structures ensure their efforts are aligned with clear, rigorous goals.

As states and districts consider changes to their accountability systems, lowering standards and misleading students and families about the reality of academic progress is not the answer. The new NAEP results make it clear that now is the time to raise expectations and work together to leverage every tool available to advance learning.

I am also the parent of four children who attend Albuquerque schools and a former fourth-grade teacher.

So, when I heard that school boards focused on students’ academic and outcome data, I was shocked. 

The reality is that school boards spend too much time naming buildings and debating sports schedules or determining which paper towels to buy.

School boards play a vital role in empowering local governance and harnessing the power of democracy to address challenges closest to home. So in 2021, I decided to channel my shock at how little attention my hometown school board paid to academic outcomes and ran for a seat. 

I knocked on 8,000 doors, and today, I’m president of the Albuquerque Public Schools Board of Education. We now spend significant time on matters related to student outcomes at our meetings. During one recent session, we looked at state assessment data and focused on unconscionably low achievement rates among Native American students. The conversation shined a light on the problem, and the district superintendent committed to doing things differently, starting with sharing this data with tribal leaders. 

Shifting conversations, and the way the board works, hasn’t been easy. Political distractions and ingrained practices like trying to manage schools instead of letting superintendents do their job get in the way.

But focusing on student outcome data has never been more important, as the recently released National Assessment of Educational Progress results show. The data is troubling.

Students are still experiencing declines in reading. Scores on the NAEP 鈥�  also known as the Nation’s Report Card 鈥� are down nationwide in both fourth and eighth grades, and that compounds declines seen on the 2022 report card. In math, scores increased in fourth grade, an area of resilience, but they’re flat in eighth grade after crashing historically the last time the test was given. 

NAEP is the only common assessment that allows policymakers and education leaders to compare student achievement across states and more than two dozen urban districts. Albuquerque is among those districts that get NAEP scores, and I’ll be studying our results with my colleagues. 

For example, I’m particularly interested in what’s going on with middle school, because our eighth-grade scores dropped in both math and reading. And I also want to know why our fourth graders posted flat scores in math while the nation overall made progress.

Of course, the work can’t just be about studying the data. We have to act on it, too. Our board has set for math, reading and college and career readiness, and we assess progress in these areas every month. We also hold our district’s leadership accountable for these goals, which ensures critical issues get the attention they deserve. For example, we鈥檝e seen grade-level proficiency grow for targeted student groups from 11.3% in 2023 to 12.8% in 2024. The proficiency rate today is 2.5 percentage points higher than in 2022 and represents real learning for students who are Native American, African American, economically disadvantaged or English learners, or who have learning disabilities. 

If you’re a school board member, or a citizen who wants to see action on these issues, here are steps you can take today.

• Read the NAEP results. These are reported for the nation, by state and for 26 large urban districts. NAEP also disaggregates data by student subgroups. As a group, Hispanic eighth graders nationwide saw the biggest declines in reading and math. It warrants asking how they are doing in your community. 
• Think about questions you want to ask your school district leaders after looking at the data. 
• Set measurable goals based on multiple, reliable data sources. Use your power, either as a board member or engaged parent or citizen, to hold your district accountable for meeting them. 
• Don’t forget about non-academic data points. The Nation’s Report Card includes student survey data on confidence and chronic absenteeism, both of which are improving somewhat but aren’t back to pre-pandemic levels. Analyze this data and see if it offers insights into the well-being of your students and the culture and climate of your schools.

As a parent, I experienced the frustration of seeing my sixth grader doing fourth-grade work during the pandemic. I could clearly see the problem because she went to school virtually from my kitchen table. Unfortunately, I wasn’t able to get school or district officials to respond to my concerns. That is among the reasons I ran for the board. From this perch, I know school board members can play a leading role in ensuring that schools are responsive to student needs and parent concerns. But they can’t do that unless they turn their attention to the things that really matter. 

However, as the most recent eighth-grade NAEP results vividly illustrate, many young people who begin the school year behind are falling even further back: The test showed a sharp decrease in proficiency levels for those at the 25th and 10th percentiles. That’s because math is cumulative 鈥� what’s learned in one year is foundational to what’s taught in the next. Miss out on key concepts in one grade (as many did during the pandemic), and learning gaps can snowball for many more.

HQIM is not designed to address unfinished learning from prior years. As a result, math educators have the massive challenge of both teaching grade-level material and addressing students鈥� individual needs. To cope, they often simplify HQIM, use less rigorous materials or abandon recommended teaching methods. These approaches both dilute HQIM鈥檚 benefits and perpetuate unfinished learning.

During regular, grade-level math lessons, otherwise known as Tier 1 instruction, some schools now use software to tackle learning gaps. While this can help, significant gaps may require additional dedicated support, called Tier 2 instruction.

Tier 2 instruction aims to help struggling students catch up, but it often misses the mark. It may replicate Tier 1 lessons without addressing gaps or focus too narrowly on basic skills without connecting them to grade-level material. This disconnect makes it harder for students to bridge their knowledge gaps.

Moreover, Tier 2 teachers face diverse student needs, with gaps spanning multiple years or skills. Without proper resources, they rely on guesswork, leading to inconsistent results. That鈥檚 why a more cohesive and structured approach to Tier 2 instruction is essential.

To ensure HQIM is effective, schools need three key elements: time, actionable data and flexible instructional content.

Effective Tier 2 instruction first requires dedicated, structured time that is properly and consistently staffed. Schools might allocate part of the core math block, additional supplemental periods or even after-school sessions. It鈥檚 also crucial to establish a team of Tier 2 instructors and promote collaboration between Tier 1 and Tier 2 teachers to align their goals and efforts.

Second, both Tier 1 and Tier 2 teachers need accurate, timely information to understand students鈥� learning gaps. Diagnostic assessments at the start of the year or before each unit can pinpoint missing foundational skills so Tier 2 lessons can connect to the grade-level topics covered in Tier 1. Once the school year begins, the use of skill-level assessments across both Tier 1 and Tier 2 can help to ensure a real-time and shared understanding of each student鈥檚 unique learning profile. If these kinds of diagnostic and skill-level assessment tools aren鈥檛 available, teachers will need to connect the results of students鈥� prior assessments to the prerequisite skills required in future curriculum units 鈥� a time-consuming but invaluable process.

Lastly, Tier 2 instructors will often require instructional content that can address students鈥� relevant learning gaps from the current or prior school years. That may include HQIM lessons from earlier grades as well as the use of high-quality instructional software that students can use independently and that is compatible with their Tier 1 curriculum.

HQIM has raised the bar for math education, replacing inconsistent curricula with rigorous, equitable standards. For too long, students were subjected to fragmented and inconsistent curricula that did little to ensure equity or rigor. HQIM has changed that narrative, setting a higher bar for what students can achieve.

However, to fully realize the potential of HQIM, the education system must evolve further. The next step 鈥� HQIM 2.0 鈥� requires integrating diagnostic data, flexible instructional content and robust support systems to meet the needs of all learners. This approach will allow schools to maintain high expectations while addressing individual student needs across both Tier 1 and Tier 2 instruction.

By focusing on these foundational changes, schools can create a more coherent and effective approach to math instruction.