社区黑料

Get stories like this delivered straight to your inbox. Sign up for 社区黑料 Newsletter

Newly released data from America鈥檚 longest-running measure of student learning have delivered a decidedly split verdict on the state of schools.

Math and reading scores from the 鈥淟ong-Term Trends鈥� edition of the National Assessment of Educational Progress 鈥� a federally administered test commonly referred to as the Nation鈥檚 Report Card 鈥� offer some of the first proof of recovery from COVID-era learning loss, with the average 9-year-old improving by 4 points since 2022. Surprisingly, those gains were driven in large measure by struggling students, who enjoyed their first major leap in several decades.聽

But 13-year-olds made no similar progress, with scores in both subjects flat or declining for virtually every demographic group. Average performance in reading for these students was no higher than in 1971, when the exam was first conducted.

The differing trajectories underline a critical split among U.S. pupils in 2026. The youngest test takers were still in preschool when COVID-19 emerged, and largely avoided the most severe educational consequences of the public health emergency. But today鈥檚 middle schoolers were second- and third-graders at the beginning of the pandemic, which led to several years of school closures and virtual instruction in many areas of the country. As this micro-generation of children proceeds through their K鈥�12 careers, they bear the scars of that upheaval.

Kirsten Baesler, who leads the Office of Elementary and Secondary Education at the U.S. Department of Education, said she was 鈥渧ery excited鈥� by the progress made by 9-year-olds, while adding that the prolonged stagnation experienced by teenagers was somewhat predictable.

鈥淭hey were in some of their most formative years of both literacy and numeracy [at the onset of the pandemic], and it was a seismic event,鈥� she said in an interview. 鈥淚t鈥檚 going to take equally seismic effort to ensure that those students are coming back to where they need to be.鈥�

Learning recession

Others placed the downturn on a timeline extending much earlier than 2020. John White, Louisiana鈥檚 former state superintendent for public instruction, argued that Wednesday鈥檚 revelations were consistent with earlier research showing that students transitioning from elementary to middle school have had 鈥渁 particularly hard decade-plus.鈥� A recent analysis from scholars at Dartmouth, Harvard and Stanford labeled the period since 2012, marked by declining achievement for all but the top students, as a 鈥渓earning recession.鈥�

鈥淲e have plenty of evidence that [a] learning recession in the middle grades predates the pandemic,鈥� said White, now serving as CEO of the educational publisher Great Minds. 鈥淵ou can imagine two compounding problems: One, a general challenge in the success that American schools are having with adolescents, and two, a pandemic that hit this group of soon-to-be adolescents particularly squarely.”

Both core subjects showed signs of the division between younger and older students. 

After seeing a 4-point boost since the last version of the Long-Term Trends test, 9-year-olds have now caught up to their performance level in reading from before COVID. Their average score is now 10 points higher, on a 500-point scale, than in 1971 鈥� if not a gargantuan leap, at least measurable upward movement. In math, while significantly lower than the pre-COVID status quo, average scores are 19 points higher than in the late 1970s.

Remarkably, growth over the past few years has been powered overwhelmingly by the students performing at the lowest levels. 9-year-olds scoring at the 25th percentile (i.e., lower than three-quarters of their same-age peers), made strides of 7 points in math and 6 points in reading since 2022; those at the 10th percentile gained even more ground, ascending by 9 points in math and 8 points in reading. That momentum flies in the face of the defining pattern of the 2010s, when only the highest-performing NAEP participants posted significant gains.

By contrast, the average performance of 13-year-olds has remained flat since 2022, and is statistically worse than in 2020. Even among those scoring at the 75th and 90th percentiles in math have endured a significant dropoff during that time.

In 2012, 85% of test takers in the older age group exceeded 250 points in math, a benchmark signaling their ability to solve one-step word problems involving addition and subtraction; in the most recent iteration of the exam, only 70% met that standard. The share of 13-year-olds scoring 250 or higher in reading fell from 66% to just 58% over the same period.

There was little variation between NAEP participants of various demographic categories, with children from various racial and socioeconomic groups generally following the same trajectories. But one notable exception related to sex: While nine-year-olds surpassed their overall results from 2022, only boys made statistically significant gains, jumping by an average of 7 points in reading and 5 points in math. Girls improved by a single point in reading and 3 points in math.聽

Drop in reading for pleasure

A few other secondary findings were drawn from a survey traditionally accompanying the exam, which generates thousands of student observations in order to construct a representative picture of their day-to-day experiences. Responses revealed that in-school attendance is still much lower than before the pandemic, with the proportion of 13-year-olds absent at least one day per month climbing from 44% in 2012 to 61% in 2025. Meanwhile, the fraction of 9-year-olds saying they鈥檇 been assigned no homework the previous night rose from 19% to 39% over the past two decades.

Perhaps most striking of all, far fewer students reported that they routinely read in their downtime. Just 37% of 9-year-olds, and 14% of 13-year-olds, said they read for fun 鈥渁lmost every day鈥� in 2025; those numbers peaked at 53% and 35%, respectively, over 40 years ago.

Education leadership consultant Julia Rafal-Baer is a member of the National Assessments Governing Board, the entity that helps design and administer NAEP. She observed that the reading results are indicative of a widespread and concerning decline in literacy that is likely linked to increased use of smartphones and social media.

鈥淲e’ve got to put real books back into kids’ hands,鈥� Rafal-Baer said. 鈥淟ibraries matter so much, and we’ve got to have adults helping kids to be curious.鈥�

The importance of the Long-Term Trends exam, she continued, lay in its consistency over time: The test has presented students with similar content, in a paper-and-pencil format, for a half-century. Even amid the education community鈥檚 often-loud debates over curriculum and accountability, the same fundamental skills have been assessed and recorded. In her view, that makes this version of NAEP 鈥渢he closest thing we have to a long-term memory of how kids are doing.鈥�

“There have been periods of time when we really did see growth,鈥� Rafal-Baer reflected. 鈥淲e were climbing for decades, and then we peaked around 2012 and have dropped ever since.鈥� 

Bringing the 鈥榗louds in鈥� 

For Eric Hanushek, a Stanford economist who sat on the governing board from 2019 to 2023, the lengthy slide in student outcomes is the central phenomenon of K鈥�12 schooling since the Obama administration. Even the apparent progress made by the youngest group of test takers has not dislodged his view that transformative changes are needed for the education system to turn things around.

鈥淓very time we see a little bright spot about what 9-year-olds are doing, for example, people jump on it as though it’s a long-run trend,鈥� Hanushek said. 鈥淚t’s going to take a lot to convince me that we’re not still in a general downhill slide, even with some nice green shoots here and there.”

A longtime skeptic of various school improvement efforts, he noted the long list of policies adopted throughout the U.S. since NAEP debuted, from increasing per-pupil spending to reducing class sizes to heightening academic accountability requirements. While some growth had been achieved, particularly in math, his assessment of the situation was largely disappointing.

鈥淚鈥檓 here to bring the clouds in,鈥� he joked.

Baesler even voiced some doubts about the validity of the Long-Term Trends assessment, noting that its testing format and some of its content could be seen as antiquated by today鈥檚 standards. The disjunction between some of the verbiage and expectations of the Ford administration and those of the Trump era may argue for an update, she continued.

At the outset of Trump’s second term, rumors circulated Washington of a forthcoming purge of NAEP exams, possibly to include Long-Term Trends. The assessment for 17-year-olds was, in fact, cancelled early last year.

鈥淭here is discussion being had鈥� about the fate of the test going forward, Baesler said.

鈥淭here needs to be serious consideration whether we should continue the Long-Term Trends, whether it is valid and accurate.鈥�

Did you use this article in your work?

We鈥檇 love to hear how 社区黑料鈥檚 reporting is helping educators, researchers, and policymakers.