Le plus ça change…

The past year marked the beginning — or re-beginning — of a new, old era in American schools: the Trump administration’s second term, which promised an explosion of school choice programs, further rollbacks on controversial content in classrooms, and a radical reduction in the federal government’s intervention in public schools. There was a new sheriff back in town.

To one extent or another, those priorities have all been embraced as predicted. But they were also the K–12 hallmarks of President Trump’s first term, however, the somewhat muddled results of which were largely overwhelmed by the chaos of COVID. Even beyond the pendulum lurches between presidencies, many of the perennial debates over education policy, politics, and governance in the United States seem to carry echoes of the distant past: Will the U.S. Department of Education cease to exist? Too bad we can’t ask Ronald Reagan.

Yet education research shows clearly how the renewed fervor of the second MAGA wave has, in some senses, fulfilled the hopes and anxieties embedded in the first. While educators that fear of immigration authorities could depress school attendance among English learners, multiple studies now persuasively link ICE and Border Patrol operations with rising absenteeism in local schools. Early evidence from states implementing voucher-like programs suggest an enthusiastic uptake among families that could have barely been dreamt of in the 2010s. And the president’s has now gone national, with county-level analyses of MMR shots revealing unmistakable downward movement since 2020. 

Indeed, this fruitful year for social science came even as the White House made good on campaign commitments to liquidate Education Department staff, cancelled dozens of contracts with research firms, and rescinded grants that had been awarded through the National Science Foundation. It remains to be seen to what extent these steps will limit the public’s insight into how schools perform and children learn, but the early signs are foreboding.

For now, though, it’s worth reviewing the empirical insights that taught us the most about education in 2025. Welcome to the year in charts.

ICE

Immigration Enforcement Worsened Absenteeism

The impact of the Trump administration’s clampdown on illegal immigration this year was felt immediately, with the number of detainees held by U.S. Immigration and Customs Enforcement between January and August. That aggressive approach is correlated with a meaningful decline in school attendance, according to released by Stanford University economist Thomas Dee.

Studying the aftermath of in California’s Central Valley — initiated a few days before President Trump was inaugurated, using tactics that would by a federal judge — Dee found that student absences increased by 22 percent across five surrounding school districts. Extrapolated across the rest of the school year, he calculated, the reduced attendance would result in over 700,000 lost days of student learning.

Similar trends were observed in on Connecticut and Rhode Island, where absences for English learners increased by 4 percent. Just as striking, showed that Spanish-speaking students at schools that were more exposed to heightened immigration enforcement experienced both lower academic achievement and fewer disciplinary incidents. 

Learning Loss

Districts Still Lag Pre-COVID Achievement

More than a half-decade after the first COVID-19 cases were detected in the United States — today’s high school seniors were just wrapping up the sixth grade when emergency school closures were announced — K–12 learning has not fully recovered in most communities. 

A February report from the , a research consortium dedicated to studying the pandemic’s effects, found that just 6 percent of American elementary and middle schoolers live in school districts where average math or reading levels have returned to the levels seen in 2019. Combining state test scores for 35 million students with results from the National Assessment of Educational Progress, a federally administered exam known as the “Nation’s Report Card,” the authors estimated that the average American pupil was still a half a grade level behind pre-pandemic achievement in both core subjects.

Even given the enormous federal funding spent to offer catch-up instruction, the prospect of improvement can’t be taken for granted: Scant improvement was seen in student scores between 2022 and 2024, with reading performance actually declining after schools reopened.

“Given 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,” Harvard economist Thomas Kane told The 74. “But no, actually.” 

Academic Achievement

K–12 Learning Was Stagnant Before COVID

The academic damage inflicted by the pandemic has been well chronicled by Kane and others. But a range of voices rose this year from around the K–12 world to critique other sources of the national school stagnation, including and . 

One, Stanford economist Eric Hanushek, circulated this fall charting the trajectory of educational outcomes in the 21st century, arguing that the years preceding 2020 were no golden era of academic achievement. According to his estimates, only one-quarter of the decline in students’ average reading performance since 2013 — often described as the high-water mark of learning, as measured by NAEP — took place during the pandemic itself. What’s more, he continued, the cost of that lost educational growth could number into the many trillions of dollars as time goes by.

It is an assessment largely shared by the equally prominent University of Virginia researcher James Wyckoff, who released looking at similar trends earlier in the year. That study asserts that the roots of learning loss began even earlier than is typically assumed, in the years spanning the Bush-Obama transition, and notes an abundance of potential explanations (including smartphones, Great Recession-related school funding cuts, and the implementation of Common Core standards). 

Readers can expect to see more such forensic examinations of the last few decades. As the shadow of COVID recedes further into memory, policymakers will need to come to grips with the K–12 fundamentals that hurt student performance in the 2010s.

ESA’s

Education Savings Accounts Lift Private School Enrollment — and Tuition

The march of private school choice picked up steam in 2025 as the availability of education savings accounts, which provide families thousands of dollars to use for private tuition or other K–12 expenses. Most were conservative bastions like Idaho, Wyoming, or Tennessee — Texas, in particular, is poised to become the largest school choice marketplace in the country — but New Hampshire also became the first blue state (albeit Republican-led, for the moment) to join the party.

With so much of the country stampeding toward the policy, social scientists are also beginning to understand its effects. In , Tulane University’s Douglas Harris and Gabriel Olivier studied the 11 states that enacted universal ESAs between 2021 and 2024, ultimately discovering that they led to slight bumps in both private school tuition (5–10 percent) and enrollment (3–4 percent) compared with other states. The bump in costs is particularly noteworthy, they argue, given that many private schools already substantially raised their prices during the COVID era.

While that study incorporated data on nearly 60 percent of all U.S. private schools, the RAND Corporation also released at the ESA picture in one state, Arizona. That work also revealed that the switch from income-targeted to universal ESAs led to a 12 percent jump in elementary school tuition; in a pattern that will likely take hold elsewhere, the number of students participating in the program also leapt from 12,000 in 2021 to almost 90,000 in 2024.

Digital Distractions

Cell Phone Bans Boost Student Performance

While the ESA wave is still building, the push for phone restrictions in classrooms has exploded to . Thirty-seven states have either passed laws to curb phone usage in K–12 schools or required school districts to adopt their own policies to similar effect. The movement caught fire over the last few years in response to complaints from families and educators that digital devices present a major distraction during the school day.

In investigating the effects of such restrictions in Florida, University of Rochester Professor David Figlio and RAND economist Umut Özek found that bans were associated with slight improvements in standardized test scores (1.1 percentiles, on average, two years after a ban was put in place) that were somewhat larger for male students and those enrolled in middle or high school. A substantial portion of that boost, they write, is attributable to improved attendance resulting from the bans.

Not all of the news was rosy. Male students, and particularly African Americans, also saw elevated rates of disciplinary infractions and suspensions in the wake of a phone ban taking effect, though that trend mostly subsided by the second year following the adoption.

Academic Standards

Exam-Free Admissions Lowered Standards at UC

The late-breaking leader for the most stunning chart of the year came in November, with the release of from the faculty Senate of the University of California San Diego. A working group convened to study the academic preparation of students reported that over 12 percent of incoming freshmen in 2025 could not meet high school math standards — a figure that had increased by a factor of 30 since 2020. An astounding 8 percent of freshmen could not perform to middle school standards in the subject.

While COVID-era learning loss is undoubtedly to blame for some of these developments, the faculty group also cast blame on slumping standards at both the K–12 and higher education levels. Among students channeled into the university’s remedial math course, which was originally designed to teach a tiny fraction of freshmen, roughly one-quarter were admitted with 4.0 GPAs. 

Those deceptive grades became a crucial indicator of student readiness in 2020, when the UC system in admissions decisions. Whatever the chief explanation for these results, one thing is clear: At an internationally recognized college, considerable numbers of students are paying tens of thousands of dollars per year to be taught material they should have mastered years ago.

Workforce

Interest in Teaching Lower Among Men, Non-Whites

A wave of recent research a pronounced decline in the prestige of the teaching profession and the job satisfaction of in-service educators. In , academics from the University of Virginia and Texas A&M put that swoon in particularly stark relief by investigating exactly who expresses interest in becoming a teacher.

Gathering data from 64 million college applications between 2014 and 2025, the authors were able to assess the aspirations and traits of high school seniors who declared an interest in the teaching profession. The group was polarized heavily on lines of sex, with males roughly one-third as interested in the career as females. Black students, similarly, were only about one-third as enthusiastic about teaching as their white counterparts. 

Even more intriguing, the paper leverages the applicants’ college recommendations to get a detailed view of how they were perceived by their high school teachers. Rated on a set of personal characteristics, potential future educators were described as being relatively higher in leadership, integrity, and care for others — but relatively lower in intellectual promise, academic achievement and self-confidence. 

Public Health

Vaccination Rates Are Lower

The politics of public health have also become more divisive of late, particularly in response to pandemic-era public health measures. This year, the ascent of Health and Human Services Secretary Robert Kennedy — who that childhood vaccines have no relationship to autism — marked a dramatic change in federal policy toward pediatric medicine.

A published in the Journal of the American Medical Association gives reason to think these fractures are leading to real consequences. Using data from 32 states, the authors calculate that vaccination rates for measles, mumps, and rubella have declined significantly since the emergence of COVID. Across roughly 2,000 counties, more than 1,600 reported declines during that period, and only four states (California, Connecticut, Maine, and New York) saw increases in the median vaccination rate at the county level. Around the nation, the average rate fell from 93.92 percent to 91.26 percent; the herd immunity threshold is 95 percent, the authors note.

“I think this is only going to get worse,” one pediatrician and vaccine advocate told The 74. “I think vaccines are under attack.”

Demographics

Enrollment Losses Are Steeper in Wealthier Schools

COVID scrambled the populations of districts across the country, with early estimates showing in total U.S. public school enrollment in the 2020–21 school year. But that exodus was seen in radically different magnitudes depending on school demographics, according to research from Boston University’s Joshua Goodman and Abigail Francis. 

Focusing on data from Massachusetts, that the state’s white and Asian enrollment figures were significantly lower (3.1 percent and 8.1 percent, respectively) by fall of 2024 than pre-pandemic trends would have predicted. Black and Hispanic enrollment, meanwhile, is now actually higher than expected — a major change in racial composition just in the course of a half-decade.

Notably, declines in public school enrollment were at their most extreme in Massachusetts’s wealthiest communities, where families were to private or charter schools that prioritized reopening for in-person instruction: The total number of students who disenrolled from the most affluent districts (those ranked in the top 20 percent of average income) greatly exceeded departing students from districts in the bottom 80 percent. 

Early Learning

Montessori Preschool Beats Other Models

The long-respected Montessori pedagogical model has gained a buzzy cultural prominence of late, as evidenced by for wooden toys in new parent groups. New evidence from the University of Virginia and the American Institutes for Research suggests that all those building blocks are well worth the price tag. 

In a study of roughly 600 students in preschool and kindergarten, the researchers pointed to large benefits from attending one of two-dozen public Montessori programs. Compared with a control group, the Montessori students (randomly selected through an admission lottery) enjoyed sizable advantages in executive function, reading, and short-term memory. What’s more, partly thanks to larger class sizes, the program cost over $13,000 less per child than conventional preschool offerings over a three-year span.

Considering the for this type of schooling, parents can likely expect to see more private Montessori options emerge in the coming years. But the authors conclude that expanding access to public programs “may be a cost-effective way to sustain early learning gains at least through the end of kindergarten.”

Education Polarization

College Majors Shift Students’ Politics

Much of the last two decades of political history have been characterized by “education polarization” — the of voters without college degrees to vote Republican while their more credentialed counterparts favor Democrats. The question is whether that phenomenon is the effect of college itself, or simply the product of ideological self-sorting.

According to , the actual content of college courses plays an important role shifting undergraduates along the partisan spectrum. Drawing on hundreds of thousands of survey responses from undergraduates at 377 colleges, the authors learned that students increase in “liberal” or “far-left” self-identification by seven percentage points between college entry and graduation. That average conceals a great deal of variation, however: Controlling for a swath of variables like pre-college ideology, life goals, and intended major, the data suggests that the act of studying social sciences or humanities makes students four percentage points more likely to consider themselves on the political left relative to those focusing on natural sciences. Majoring in economics or business, meanwhile, decreases that likelihood by six points. While all college students tend to slide leftward in their cultural politics, economic issues like taxation are particularly sensitive to major choice.

In all, the influence of enrolling in social science or humanities coursework over voting preferences is about equal to the effect of growing up in a heavily Democratic congressional district. If all students in left-leaning disciplines switched to business or economics, the paper estimates, education polarization would decline by roughly one-third.

Critical Race Theory

CRT in Classrooms Isn’t a Myth

The debate around critical race theory in schools has raged for much of the last half-decade, with many conservatives alleging that children are bombarded daily with messages derogating American history and Western values. Defenders of public education have responded by calling CRT an obscure sub-discipline of legal education with little purchase for K–12 students.

In January, published in the journal Education Next offered some evidence that CRT — or, at least, some of the key ideas proliferated by its academic theorists — does indeed find its way into high school classrooms. More than one-third of respondents said their teachers characterized the United States as a racist nation “often” or “almost daily,” while similar proportions reported hearing frequent messages about the racist complicity of white people and police officers. At the same time, majorities of pupils also said they’d been taught that the country had made strides toward racial equality since the 1970s.

It is difficult to say with certainty which areas see more of this kind of teaching, or even in what context such statements are made. “I’m sure there are schools where it’s not happening at all,” University of Missouri Professor Brian Kisida told The 74. “I’m also sure that there are schools where it’s happening quite a bit, and it’s really ingrained in the approach that those schools take.”

Nearly five years after the emergence of COVID, the pandemic’s after-effects still ripple through schools and communities, with student learning persistently failing to reach levels seen in 2019. Just under $200 billion in federal assistance to states, which was used to keep districts afloat during the crisis, expired in September — with no further help visible on the horizon.

Increasingly, though, the kids filling American schools have only dim memories of quarantines or virtual instruction. Their experience is instead defined by a rash of trends and technologies that sprang up, or became much more common, during the period when schooling was scrambled: a massive build-out of tutoring programs; the rapid adoption of artificial intelligence as a tool of both academic achievement and academic dishonesty; a rise in student despair and anxiety, which some experts attribute to the spread of smartphones; and, for adolescents, soaring recreational marijuana use under newly permissive state laws.

Tomorrow is coming faster than ever, and its contours will be shaped by new leadership in Washington. After a fervid campaign season, President-elect Trump has already vowed to essentially terminate the federal government’s role in setting education policy by eliminating the U.S. Department of Education. 

But before turning to the future, The 74 is taking a look back at 2024’s biggest discoveries from the world of education research. Welcome to the year in charts.

Federal Funds Lifted Learning — But Not Enough

Two papers released this summer by the Education Recovery Scorecard and the Center for Analysis of Longitudinal Data in Education Research attempted to quantify the effects of the federal government’s , which channeled $190 billion to schools and districts over the last four years in response to the pandemic. Their findings showed that the money has helped, but came nowhere close to filling the academic hole left by COVID.

ESSER’s benefits were relatively modest (measured in math test scores, each $1,000 spent yielded about 10 percent of what is generally considered a medium-sized effect in education research) and distributed unequally, as different school districts received wildly divergent amounts from Washington. Assuming a similar bang for the buck, Congress would have to appropriate between $450 and $900 billion in further legislation in order to bring learning back to where it was in 2019, the researchers estimated.

That’s almost certainly not going to happen; ESSER funds officially dried up this September, and no effort has been made to renew them. If no further assistance is coming, the program’s legacy will have been helping to spur an incomplete learning recovery: According to released by the leaders of the Education Recovery Scorecard, students across the country had only made up one-quarter of their lost progress in reading, and one-third of their deficits in math, by the beginning of this year.

Students Are Still Hurting

The full picture of learning loss remains discouraging, particularly for those who were in their foundational years of schooling when the pandemic threw their education into chaos. 

According to by the testing group NWEA, eighth graders in 2024 were still a full school year behind in both math and reading compared with similar students from five years prior. Derived from the scores of 7.7 million students on the organization’s MAP Growth measure, that assessment also pointed to racial achievement gaps that have only grown wider in the 2020s, with Hispanic students falling the furthest behind in both elementary and middle school.

While academic damage has been especially scarring for those in middle and high school, even elementary schoolers are making slower academic progress today than in previous years. A separate report, released in March by the curriculum provider Amplify, showed that students from kindergarten through the second grade are making less progress toward literacy than they did during the 2021–22 and 2022–23 school years. In other words, growth has even slowed down since the immediate post-COVID period.

The Disappearing College Freshman

Colleges and universities face punishing demographic challenges in the years to come, as smaller birth cohorts and shrinking high school classes leave institutions to fight over a diminished applicant pool. Even more worrying, data suggests that rising numbers of potential college-goers are reconsidering their future plans and heading . 

The end result is a surprising erosion in the numbers of rising college students. According to by the National Student Clearinghouse Research Center, freshman enrollment has declined by 5 percent since last year, with 18-year-old freshmen falling by 6 percent. What’s more, that drop comes after a 3.6 percent decline just last year.

Much of the shrinkage was concentrated in particular student demographics and institutional types. For example, the number of white students — who constitute a healthy majority of all college attendees — fell by 0.6 percent this year, while their non-white peers continued to tick upwards. Most striking of all, both public and private colleges that enroll high percentages of Pell Grant recipients saw double-digit losses in freshman enrollment. 

Charter Schools Boost College-Going, If Not Test Scores

Charter schools have long enjoyed an uneven reputation based on geography. While those located in cities — often built on a “no excuses” framework that emphasizes high standards and tough discipline — can achieve incredible results, their suburban and rural counterparts traditional public schools.

But a paper authored by University of Michigan researcher Sarah Cohodes added a striking addendum. In an experiment based in Massachusetts, where Boston-based charters post anywhere in the country, she discovered that non-urban charters also manage to significantly increase students’ chances of enrolling and graduating from college. Paradoxically, however, they do so even as those same students perform worse on standardized tests than their peers in nearby public schools. 

It’s an open question how children’s achievement could decline even as post-secondary outcomes improve. Cohodes allowed for the possibility that families in suburban and rural school districts might enroll their kids in charters that focus heavily on areas like arts programming or social-emotional instruction, rather than elevating achievement in core subjects like math or English. 

“The whole premise of test-based accountability is that test-scores predict longer-term outcomes,” Cohodes told The 74. “But this situation shows it is not always the case, and other things are going on in schools.” 

AI Could Get the Most out of Tutors

Tutoring programs exploded in the last five years as states and school districts searched for ways to counter plummeting achievement during COVID. But the cost of providing supplemental instruction to tens of millions of students can be eye-watering, even as the results seem to taper off as programs serve more students.  

That’s where artificial intelligence could prove a decisive advantage. circulated in October by the National Student Support Accelerator found that an AI-powered tutoring assistant significantly improved the performance of hundreds of tutors by prompting them with new ways to explain concepts to students. With the help of the tool, dubbed Tutor CoPilot, students assigned to the weakest tutors began posting academic results nearly equal to those assigned to the strongest. And the cost to run the program was just $20 per pupil. 

The paper suggests that tutoring initiatives may successfully adapt to the challenges of cost and scale. Another hopeful piece of evidence appeared this spring, when Stanford University researchers found that a “small burst” program in Florida produced meaningful literacy gains for young learners through micro-interactions lasting just 5–7 minutes at a time. If the success of such models can be replicated, there’s a chance that the benefits of tutoring could be enjoyed by millions more students.

Teachers Aren’t Happy

K–12 educators have had a tough few years. While there’s strong disagreement about just how many of them actually walked off the job during the worst years of COVID, a combination of public health fears and worsening conditions in schools has led many to consider leaving the field since the pandemic began.

A published this fall by Brown University economist Matt Kraft put those fears into a much larger context. Using polling data going back decades, he found that public esteem for teaching — as measured by how many people called it a prestigious career, compared with other professions — is now at the lowest level seen in half a century. Fewer than half of all teachers said that the stress of their job was worth the effort, compared with over 80 percent in the 1970s.

Those numbers are bad enough, but they also appear to be turning off potential teaching candidates. The number of newly licensed teachers fell by one-third between 2006 and 2020, indicating that the reputational problems facing the K–12 workforce came about long before the pandemic. Interest in teaching as a career path among high school seniors and college freshmen has also dropped substantially since 2010.

Even with a precipitously shrinking number of K–12 students, schools will have a hard time coping if this generation of educators is replaced by a smaller, more demoralized cohort of successors.  

The Culture Wars Are Coming to a School Near You

One likely reason for lower job satisfaction among those toiling in the classroom? Disputes over politics and culture, which have recently grown far more contentious.

released by the RAND Corporation in February first publicized what many school employees have complained about for years. Lawmakers in 18 states passed legislation restricting classroom discussion of some topics, whether related to politics, history, race, gender, or sexuality, between 2021 and 2023. Those states are home to approximately one-third of all American teachers.

Strikingly, however, a full two-thirds of all teachers polled by RAND said that they self-censored or otherwise curtailed dialogue with students about hot-button issues. The authors dubbed that trend a “spillover” between school communities, often driven by groups of particularly vocal parents who may not reflect the attitudes of their neighbors. In the end, more than half of all teachers working in states with no statutory restrictions on classroom discussion still self-censored to one degree or another, the poll indicated. 

Notably, those findings dovetail neatly with other research showing that clashes over culture war issues can be and potentially harmful to student learning. 

Screentime Is On the Rise. So Is Depression

This year will likely be remembered as the period when concerns over children’s smartphone use, both inside schools and out, came under a microscope as never before. An increasing number of schools in the United States and around the world have moved to restrict the use of phones in the classroom, with many complaining of both disengagement during lessons and an atomized culture brought about by technological distraction.

But a growing scientific literature suggests that young people may be profoundly impacted by phones and social media during their hours at home and with friends. In , British academic Danny Blanchflower — a labor economist who has also specialized in the study of public happiness over decades — demonstrated a close correlation between the steep increase in youth exposure to screens and a concurrent upswell in self-described feelings of despair, worry and self-doubt. 

In 2022, Blanchflower and his colleagues found, over one-in-ten young women said they’d experienced a bad mental health day every day over the previous month, tripling the rate they’d reported in the early 1990s. At the same time, the percentage of young women who absorbed more than four hours of screen time each day jumped nearly eightfold.

Arguments about the effect of information technology on youth mental health are hotly contested, with skeptics observing that the evidence for a firm casual relationship between smartphones and depression is still quite tentative. But Blanchflower believes the downside risk of unfettered screentime is too great for policymakers not to act.  

“We could fart around about causality, but the potential cost of not doing something is so much greater than the cost of doing something and being wrong,” he told The 74.

Catholic Schools Might Need Vouchers to Survive

Since the beginning of the charter school explosion in the late 1990s, denizens of the policy world have speculated that the birth of a new educational model could escalate the decades-long decline in Catholic schooling. While increasing secularization has likely driven much of , the more recent emergence of free, easily accessible schools of choice in virtually every major American city seemed like the equivalent of throwing an anvil to a drowning man. 

In , Boston College professor Shaun Dougherty offered persuasive evidence that charter expansion had indeed come at the expense of the Catholic sector. Relying on data collected from over 25,000 K–12 institutions, the study calculated that between 1998 and 2020, an average of 3.5 percent of Catholic school students disenrolled within two years of a charter opening in the vicinity. Given the thin margins in Catholic education, those declines made full-on closures significantly more likely. 

In a telling wrinkle, those trends were considerably muted in 10 jurisdictions that offered some form of private school choice, which provides families with money to spend on tuition or other educational expenses. That suggests that, with the spread of education savings accounts and similar policies, the multi-generational eclipse of Catholic schooling may begin to slow or even reverse. But, as Notre Dame law professor Nicole Stelle Garnett told The 74, it could be too late for the Church to reverse its losses.

“If we’d gotten this much of private school choice in 1999, instead of 25 years later, we might have a lot more kids in Catholic schools today.” 

School’s In, So Is Crime

As community hubs attracting large numbers of young people, schools are somewhat unavoidably linked to violence and antisocial behavior. Previous research has shown that when low-performing schools in Philadelphia were permanently closed in the early 2010s, the surrounding areas saw a pronounced reduction in violent crime.

But released this fall gave a much more sweeping overview of the link between schools and disorder. Using data from the National Crime Victimization Survey, the authors found that criminal activity among children from the ages of 10 to 17 — whether as perpetrators or victims — peaks during the school year, particularly during the autumn and spring. That’s an exact inversion of the pattern for older offenders, who are much more likely to commit crimes during the summer months.

Across more than 3,000 school districts, the school calendar was linked to a 41 percent increase in youth arrests and a 47 percent increase in reported crime, with the surge mostly occurring during school hours and during the week rather than the weekend. Much of the lawbreaking even occurs in schools themselves. 

“In poor and rich counties; well-resourced school districts and poorly resourced school districts; and rural and urban counties, schools are a primary driver of criminal activity involving children,” the authors conclude.

For High Schoolers, Weed is Everywhere

One form of vice is particularly prevalent among older adolescents: marijuana use. According to published in March, over 30 percent of seniors reported using weed over the past year. 

That figure reflects a few coalescing trends, most importantly the legalization (or decriminalization) of weed . Three-quarters of Americans now live in a jurisdiction where the drug is available for either medicinal or recreational use, though age restrictions still make it illegal for almost any high schooler to do so legally. What’s more, the development of kid-friendly gummies and vape flavors makes marijuana more accessible to young people than in decades past. 

That’s especially concerning given the elevated potency of new cannabis items, which are far stronger on average than the common street product of even a few decades ago. Youth marijuana use to inhibited brain development and increased risk of psychological disorders in later life.

“The biggest consequence that we think about in the field of child development … is that using substances that are potentially psychoactive and addictive and have effects on development,” Columbia psychiatrist Ryan Sultan told The 74’s Amanda Geduld. “The younger you are, the more problematic they might be.”

Pre-K Helps Families’ Bottom Lines

Early childhood education has been shown to be an effective tool for improving students’ near-term academic performance, though research is unclear on can be sustained over time. In the hopes of reaching students before the K–12 years and combatting gaps in readiness and achievement, a growing number of states and cities have their public pre-kindergarten offerings in recent years.

A paper released in October found that one such expansion brought considerable benefits to participating families — but for a somewhat surprising reason. When New Haven, Connecticut, established a pre-K program in the 1990s, enrolled students saw only ephemeral improvements to their test scores, school attendance, and likelihood of being held back in school, with effects essentially disappearing by the time they finished the eighth grade. But by participating in the program, which provided 10 hours of instruction and supplementary programs each day, those children allowed their parents to work more during the day. On average, caregivers earned 22 percent more, or nearly $5,500 per year for each year their kids remained in pre-K.

Even better, the same parents went on to earn 21 percent more in the six years after the program ended, likely because of their increased experience and job continuity, and their higher income dwarfed the costs of implementing the program. In other words, even if it contributes little in long-term academic gains, pre-K may generate huge value purely as a childcare benefit.

According to the  national results, low-income fourth graders read an average of two to three grade levels below their higher-income peers. 

It’s not new that students in poverty have lower scores on reading tests than more affluent students. Housing prices, parent perceptions and online school ranking websites all focus on those raw, unadjusted scores, which ignore the fact that some schools and districts simply have a harder job. 

But poverty is not destiny, and some schools and districts hugely outperform what might be expected of them based solely on which students they serve. 

Working with Eamonn Fitzmaurice, The 74’s art and technology director, I set out to find districts around the country that succeed with the students they actually serve. We calculated each district’s expected reading proficiency rate, based on its  rate, and compared it to its actual third grade reading scores. This methodology helped us identify districts that are beating the odds and successfully teaching kids to read.  

Select from the menu below to find the high fliers in your state. 

INTERACTIVE

Third Grade Reading Proficiency

Steubenville City, in the Rust Belt along the very eastern edge of Ohio, topped our rankings. It has a very high poverty rate — greater than 96% of districts nationally — yet 99% of its third graders were proficient in reading last year. (For more on how Steubenville achieves such impressive results, see this 2012  and this .)

Every state has its own pockets of success (represented as gold circles in the graphs). These “exceptional districts” are in the top 5% of their state, in terms of outscoring their expected reading proficiency rate. For example, Worcester County in Maryland serves about 7,000 students along the Atlantic coast. Worcester falls in the middle of the pack in terms of poverty, but it has by far the highest third grade reading proficiency rate in the state.

We found positive outliers in every state. Among these are some higher-income districts like Maryville, Tennessee; Mountain Lakes, New Jersey; and Bainbridge Island, Washington, that are exceeding already lofty expectations. They also include lower-income communities, like Dearborn, Michigan, and Neshoba County, Mississippi, that are helping students achieve results that — although maybe not high in absolute terms — should still be considered achievements, given the poverty the schools and students are facing. 

In some states, poverty has more of an effect than it does in others. Given the correlation between income and test scores, readers might assume that every state’s graph looks like Rhode Island’s, where poverty is highly correlated to district reading scores and districts are tightly bunched around those expectations. 

The diagonal line in the graph is called the “best fit” line. It is meant to go through the middle of all the points on the graph, and the closer the points are to the line, the stronger the correlation is. After Rhode Island, Connecticut, Alabama, Massachusetts and Alaska have the strongest relationship between a district’s poverty rate and its third grade reading proficiency. 

But not every state has such a tight relationship. For example, contrast how tightly districts are bunched around the “best fit” line in Rhode Island with the same graph (below) for Virginia. In Virginia, the relationship between poverty and reading scores is much weaker.  

States like Nebraska, West Virginia, Kentucky, North Dakota and especially Nevada have weaker relationships between district-level poverty and reading outcomes. The number of districts a state has, how students are sorted across districts and differences in the state tests themselves can all affect this relationship.

But without controlling for poverty, a “good” school district may receive credit for student learning that it actually had little part in. This issue is especially misleading in reading. Unlike math, where learning is more closely tied to school-based instruction, reading skills are multi-faceted, and they’re more closely tied to language skills and background knowledge that children acquire at home. 

As a result, some wealthier districts may show high (raw) reading scores even though their students are picking up their skills at home — or, worse, from private tutors that families with means are able to afford out of their own pockets. Meanwhile, districts doing a good job serving low-income students have a harder time showing the same proficiency rates. But some truly are beating the odds at helping kids learn to read, and their leaders deserve praise and celebration. 

Methodology and Limitations: The data for this project come from two sources. Poverty rate data comes from the 2022 district-level figures from the . Third-grade reading scores were downloaded from , an initiative from ParentData.org and Brown University to compile state test scores and make them  publicly available. 

Because the poverty rate data from SAIPE is reported at the district level, we could not look at results for individual schools. The data also do not include standalone charter schools, so these are included when they are part of a district but not when they are considered their own district. Similarly, the data provide only one poverty rate for all of New York City, so readers should interpret those results with caution. 

We limited our sample to districts with at least 30 test-takers in spring 2023. Because different states use different tests, we encourage readers to focus on within-state comparisons only. For example, we could not include places with only one district (e.g. Hawaii and the District of Columbia). Vermont had not reported its 2023 district-level proficiency rates by the time of publication, so it is excluded. Maine did not break its results down by grade level, so its numbers use an aggregate across grades 3 to 8. All told, we had comparable data for 9,605 districts across the country.

Schools had one foot in the shutdown era, still struggling to restore a sense of normalcy that disappeared in 2020. A steep rise in behavioral and disciplinary issues, which many teachers hoped would be only the temporary product of COVID’s generational disruption to routines, stayed with us. Millions of kids have remained separated from their local schools — not because they’re prevented by public health measures from entering the building, but because they’re simply choosing not to attend classes. And across a whole range of academic subjects, actual student learning is lower and slower than it was before the pandemic.

Meanwhile, school systems are adapting to trends and technologies that have arisen just over the past few years. Districts are spending billions of dollars to establish or expand tutoring programs, which may be America’s best tool to combat learning loss, while AI platforms like ChatGPT are transforming the way instruction can be delivered (and challenging schools’ ability to keep ahead of cheating). 

And researchers continue to ask all the questions that have traditionally set the parameters of America’s K–12 agenda: Why do student populations self-segregate? Is it better for kids to be assigned to tough or easy graders? How much do teacher training programs really help? Have charters caught up to traditional public schools?

As we do every year, The 74 has compiled a year-end inventory of the most fascinating discoveries, insights, and ambiguities that came out of education research in 2023.

Welcome to the year in charts.

Student absenteeism is out of control

You could spend a lot of time simply tallying the aspects of student life that COVID made worse: significantly diminished achievement, lower odds of graduating on time, escalating behavioral challenges, and fewer applications to college. But the most dangerous consequence might be its effects on how often children came to school.

According to by Stanford University Professor Thomas Dee, the proportion of K–12 students who were chronically absent — i.e., who missed 10 percent or more of the school year — nearly doubled during the pandemic, vaulting from 14.8 percent in 2019 to 28.3 percent in 2022. Extrapolated across all schools, that means an additional 6.5 million kids became chronically absent following COVID. Every state Dee studied saw an increase of at least 4 percentage points, but those with higher pre-pandemic rates of absence experienced the largest jumps.

The findings jibe with those of other alarming research on attendance. from Johns Hopkins University’s and the advocacy group Attendance Works, covered by The 74’s Linda Jacobson in October, showed that in 2021–22, two-thirds of American students attended a school where at least 20 percent of students were chronically absent. In over half of all high schools, chronic absenteeism rates topped 30 percent that year. 

Catch-up learning hit a wall last year

But are kids (at least, the ones actually showing up) regaining the ground they lost since 2020? According to much of the testing data that emerged this year, the answer is no — or at least, nowhere near quickly enough.

In , researchers from the nonprofit testing organization NWEA combed through nearly seven million children’s scores on the , which is administered both in the fall and the spring to measure how much students learn during the year. But test takers in the 2022–23 academic year made markedly less progress in key subjects than comparable elementary and middle schoolers who sat for the exam before the pandemic, with growth in reading and math falling by as much as 19 percent and 15 percent, respectively. Only third-graders exceeded the pre-COVID learning averages. 

The stalled momentum was directly cited in on “State of the American Student,” which distilled a host of worrying trends and warned that America has little time left to reset the trajectory for millions of adolescents. According to ongoing indicators like the National Assessment of Educational Progress, which released long-run scores for 13-year-olds this spring, average performance in math and reading has been set back to levels last seen decades ago.

Even if schools and families feel like they’re through with the pandemic, the pandemic — and the harsh blow it has dealt to kids — isn’t done with us. 

Virtual tutoring can work

Thankfully, states and districts aren’t sitting on their hands in the face of learning loss. Supported by billions of dollars of federal funds, many have invested heavily in tutoring programs that promise to help struggling children overcome the challenges imposed by past school closures and virtual instruction. The question is whether those efforts work for enough students to justify their cost — and according to data generated by , a Stanford initiative devoted to studying the effects of tutoring, there is reason for hope.

In October, the Accelerator circulated showing impressive results from , a fully virtual program provided to developing readers. The study found that among 1,000 students enrolled in Texas charter schools, participating in OnYourMark resulted in kindergartners gaining the equivalent of 26 extra days of learning in letter sounds and first graders receiving 55 additional days of sound decoding. The news is particularly encouraging in that it shows a path to success for virtual tutoring, which has often been shown to be far less effective than in-person instruction. 

Grade inflation got worse during the pandemic

As the chaotic transition to online learning got underway in 2020, schools had to decide how they would judge the work of students cut off from their teachers and classmates. Many opted for , including and granting credit for , out of a desire to avoid more punitive measures during a crisis. 

It’s difficult to chart the average impact of the shift across thousands of school districts, but the Center for Analysis of Longitudinal Data in Education Research (CALDER) recently released focusing on a decade of student records in Washington State. The picture was stark: While the average middle and high school GPA for math rose by 0.11 points between 2011 and 2019, it got a boost three times that size — one-third of a GPA point, or about the difference between a C-plus and a B-minus — between 2019 and 2021. 

In general, wrote CALDER director and American Institutes for Research vice president Dan Goldhaber, the relationship between student grades and their scores on state standardized tests “has diminished over time,” particularly in math. A similar pattern is suggested by , which show scores remaining largely flat in recent years even as students’ self-reported high school grades have climbed. And just like with price inflation, GPAs that soared during the pandemic still haven’t fully come back to earth.

Tough grading has its advantages

So what are the effects of higher course marks? Several papers released this year indicate that they can be surprisingly negative.

In this fall, a trio of researchers explored the consequences of a statewide switch to more lenient grading standards undertaken in North Carolina  in 2014. The policy was meant to make grades more comparable between school districts, but in effect, it also lowered the threshold for each letter grade in high schools. It also seemed to affect various student groups quite differently. As expected, the highest-achieving kids received higher grades (though only in their freshman year), but disturbingly, struggling students didn’t receive a similar bump. They also seemed to disengage from school, accruing substantially more absences than students who weren’t exposed to the looser standards; over time, those absences likely hurt their learning, as measured by relatively lower scores on the ACT.

If easier grading holds the potential to hurt attendance and widen achievement gaps, the opposite may also be true. In a study that also focused on North Carolina schools, American University Professor Seth Gershenson discovered that eighth and ninth graders assigned to math teachers with relatively tougher grading standards later saw higher math scores throughout high school. And far from validating fears that hard classes make kids tune out, those students were also less likely to be absent from class than their peers. 

COVID hit social studies too

Much of the concern over learning loss is focused on weakened performance on the core disciplines of math and reading. In fact, the academic harm was widely dispersed. 

The National Assessment of Educational Progress — a federal standardized test often called the Nation’s Report Card — only measures proficiency in social studies every four years. The exam’s latest results, revealed in May, showed that eighth graders’ average history scores fell by five points; civics scores fell by two points, the first decline in the history of the test. All told, the results for both have fallen to levels last seen in the early 1990s, the latest evidence that COVID has triggered a generational reversal in knowledge acquisition.

The swoon came amid a national debate over how to teach about American history and government, with states like Virginia initiating significant overhauls of their academic standards. But the phenomenon appears to be international in scope: Results from , which tests over 80,000 eighth graders across 22 industrialized countries on civic knowledge, showed that large numbers of test takers couldn’t answer questions about election fairness or democratic governance. Only 55 percent of respondents said they felt their nation’s governmental system “works well.”

Choice might be good for public schools

The explosive growth of school vouchers and education savings accounts, which allow families to spend public funds on private education, has dominated the school choice debate this year. Public school choice (i.e., charters and open enrollment policies), while also controversial, has receded somewhat from conversation.

But indicates that, in addition to providing more instructional options to families that want them, intra-choice can improve learning throughout wider communities. University of Chicago economist Christopher Campos and data scientist Caitlin Kearns scrutinized Los Angeles’s , which allows families within designated neighborhoods to select among multiple high schools rather than send their children to the one nearest their home. Participation in the program, they learned, significantly increases students’ English exam scores and boosts their enrollment rate at four-year colleges by 25 percent. Those gains were concentrated among schools exposed to the most competition and those that previously performed the worst, strongly hinting that inclusion in the Zones pushed them to hold onto students by improving their offerings. 

A in North Carolina yielded broadly similar results, though with caveats. Focusing on the state’s decision to lift its cap on charter schools in 2012, the paper’s authors revealed that the move incrementally improved public schools’ value-added scores as measured by state standardized tests; that improvement, while small in scale, generated huge value in the aggregate, as the study concluded that the average public high schooler’s lifetime wages were lifted by $1,500 by allowing more charters to open. As in the Los Angeles study, the promising effects seem to have come about through competition for students.

Dispiritingly, however, the impact on pupils who actually enrolled in the charter schools after the cap was lifted was negative, perhaps because the newly established schools tended to employ more “non-traditional” models (e.g., project-based or experiential learning, such as Montessori) that weren’t as successful as existing charter options.

No one said this stuff was simple. 

Charters aren’t underperforming anymore

Charter schools have been around for over 30 years. For most of that time, their advocates and detractors have argued passionately over just how effective they really are at improving academic achievement. The primary arbiter of those disputes, most often, has been Stanford’s (CREDO), which has released over more than a decade comparing the performance of charter students with those enrolled at district public schools.

In the first few editions, those reports showed the newer schools lagging behind their traditional counterparts — evidence that the sector’s opponents’ throughout the fierce school reform battles of the Obama era. But — CREDO’s first national evaluation in a decade, including data on 1.8 million students across 31 states and cities — calculated that charter students receive the equivalent of 16 extra days of learning in literacy, and six extra days of math, than students at the local public schools they would have otherwise attended. The edge, while decidedly slight, masks larger variation among subgroups: Black students gained an average of 35 extra days of reading growth and 29 extra days of math, equal to more than a month of supplemental instruction.

Not all charters are created equal, however. published last month in the journal Education Next, and covered by The 74’s Greg Toppo, compared the performance of charter sectors in each state based on their students’ performance on NAEP. Somewhat surprisingly, the state with the top showing was Alaska, where charter students score an average of 32 points higher on the test than the national average for charter school students. Their peers in Pennsylvania, Oregon, Michigan, Tennessee, and Hawaii weren’t so fortunate, with each scoring at least 21 points lower than the national average.

Teacher prep can be rethought on the fly

Starting in spring 2020, Massachusetts launched a grand experiment: Concerned that the tumultuous working conditions of the pandemic would discourage young people from becoming teachers, the state began issuing emergency credentials to teaching candidates even if they hadn’t completed the necessary coursework to be licensed. Over the next three years, almost 20,000 such licenses were granted to instructors who worked full-time while simultaneously working to meet their licensure requirements.

Boston University’s Wheelock Education Policy Center has followed the progress of those early-career teachers. Their analysis, laid out , presents a quietly stunning observation: As measured through a combination of school-level performance evaluations, principal questionnaires, and student scores on standardized tests, the emergency-licensed teachers perform similarly to their colleagues who completed traditional teacher preparation programs. Students assigned to them were not disadvantaged in learning in spite of their unconventional path to the classroom. What’s more, by the program’s second year, one-quarter of emergency licensees — vastly more than the statewide average in Massachusetts.

The notion that aspiring educators can thrive in the profession without reaching it through the traditional channels isn’t a new one; Teach for America and other alternative credentialing programs have existed for decades, during that period. But the Massachusetts experience illustrates some of the specific benefits of dropping licensure requirements during a crisis. Namely, making entry more flexible (and shaving off the years of study and thousands of dollars in tuition that often act as a deterrent to otherwise qualified candidates) can produce a more diverse and no less effective workforce.

More good news on third-grade retention

Legislation around the science of reading has swept through dozens of states over the last decade. In part, the political success of the new literacy agenda is due to the popularity of most of its planks: evidence-backed curricula, teacher coaching, and additional resources for kids and schools that need them.

By contrast, third-grade retention — holding back students for a year if they’re not on track to succeed by the end of — plays the role of the bad cop. In spite of the existing evidence that struggling elementary schoolers in states like and can see large benefits from repeating a grade, many parents and teachers still consider that step too punitive.

But according to , the upsides of the approach extend in some unexpected directions. In a study of 12 large school districts in Florida, which has had a retention policy related to reading scores for over 20 years, researchers found that third graders made significant gains in scores for both math and reading after being held back. Even more promising, targeted students’ younger siblings also saw larger learning gains than the brothers and sisters of comparable students who weren’t retained. 

It’s unclear what feature of Florida’s law led to the positive “spillover effects,” but study co-author Umut Özek told The 74 that families might be responding in an advantageous way to the experience of their older children. “When you get a signal that says, ‘Your kid is not performing at a level that will allow them to be promoted to fourth grade,’ that’s a very clear signal that will likely induce a response from parents.”

Asian students in, white families out

“White flight,” as it’s usually understood, refers to the phenomenon of working- and middle-class white families decamping from inner cities in the 1960s and ‘70s as a response to increased crime, deteriorating local economies, and growing numbers of African American residents. It’s a , but many in the education policy world blame it for contributing to school segregation and shrinking the tax base of urban school districts.

This year, applied the concept to a different setting. A at the city level, the Princeton economist studied the movement of Asian-American students into 152 California school districts, all of them suburban and relatively affluent. The sizable growth over the decades of the early 21st century appeared to generate its own version of white flight — more specifically, for every Asian student who enrolled in local schools, 1.5 white students left. 

The departures weren’t correlated with any other demographic changes. But accompanying survey evidence convinced Boustan and her collaborators that they also likely weren’t triggered by racial animus. Instead, they pointed to white parents’ wariness of academic competition with Asian-American kids, who in virtually every academic metric. 

“Someone is showing up in the district who scores better than they do,” Boustan said in an interview with The 74. “In relative terms, the white kids are generally falling behind.”

Extracurricular activities show large racial gaps

The most significant education development of 2023 may well have been the Supreme Court’s 6-3 decision in Students for Fair Admissions v. Harvard, the case that prohibited the use of racial preferences in college admissions. The end of affirmative action as we’ve known it, occurring just as colleges like the SAT and ACT, means that admissions decisions will increasingly be made on the basis of other parts of the application package.

One of those will undoubtedly be extracurricular activities — the menu of clubs, productions, athletics, and volunteer opportunities that high schoolers have learned to embrace in order to be considered well-rounded. But if their aim is to foster diversity while adhering to new legal constraints, colleges might think twice before relying on them too heavily. drawing on nearly 6 million college applications from the 2018–19 and 2019–20 admissions cycles, participation in extracurriculars is surprisingly race-specific. White, Asian-American, and wealthy students, along with those attending private high schools, reported engaging in many more activities than their African American, Latino, American Indian, and low-income classmates. The activities they choose also tend to feature more leadership roles and confer more honors, both of which could help win a university slot.

If race, test scores, and extracurriculars are reduced in prominence, however, it’s difficult to say what will take their place. Separate campaigns have been waged against the use of admissions essays, to favor wealthier students, and , which often leverage social capital that disadvantaged kids don’t have. In the end, admissions officers might be left throwing darts at the wall.

Flexible pay has unintended consequences

The Act 10 legislation, by Wisconsin Republicans, ignited one of the most furious school reform controversies of its era. By stripping teachers of the right to collectively bargain over salary schedules and benefits, then-Gov. Scott Walker dealt a massive blow to teachers’ unions, perhaps the most influential progressive force in state politics. It was also a provocation that some credit with catalyzing the revived organizing movement of the last half-decade, which has seen a rash of teacher strikes and renewed hostility to other planks of the reform agenda.

In a study published in the education journal Education Next, Yale economist Barbara Biasi looked at the transformative effects of Act 10 on teacher labor markets, which suddenly became much more flexible as schools could opt to pay different salaries to teachers on the basis of either career tenure or classroom performance. That had some positive effects for individual districts: Younger, more effective teachers were able to win large pay increases by moving to areas where their lack of seniority wasn’t held against them.

But the state also saw an unpalatable side effect. In part because younger female teachers are more reluctant than their male counterparts to negotiate aggressively for higher pay, flexible-pay districts also saw a newfound gender wage gap begin to open. Though small on average, Biasi found that the cumulative effect over a teacher’s career could amount to an entire year’s pay.

Gifted education does little to increase segregation 

The last few years have brought a clash between advocates for educational equity and proponents of gifted education. That battle — over gifted programs’ place in the K–12 portfolio, and whether all kids truly have access to them — has largely played out in major urban districts like New York and San Francisco, where both and have been criticized for their disproportionately tiny number of seats offered to Hispanic and African American pupils.

But several studies recently emerged that tell a different story. , published in Education Next by Williams College economist Owen Thompson, examines the effect of K–6 gifted programs on the racial makeup of kindergarten and elementary classrooms. Examining enrollment information for nearly 47,000 public schools around the United States, Thompson found that the special sections are disproportionately made up of white and Asian students. But because they are so small in scope, they make a negligible impact on the overall demographics of the schools in which they are housed. In fact, eliminating every such program would not significantly change the exposure of different student groups to one another.

That doesn’t necessarily mean that gifted learning opportunities can’t be made available to more kids, however. And , by NWEA researchers, suggests that the key to welcoming more English learners and students with disabilities into accelerated classrooms is for states to enact formal mandates related to the provision of gifted services, require districts to maintain their own formal gifted plans, and regularly audit them for compliance.