Welcome to Chalkbeat Ideas, a new section featuring reported columns on the big ideas and debates shaping American schools. Next week we have an online event with Education Department official Lindsey Burke. You can sign up to join here.
It’s hardly breaking news that American students are behind academically from where they used to be.
But the specifics can get lost in a haze of headlines and data points. To start the new year, I wanted to level-set on where we stand on learning loss and recovery — to unpack what we know and don’t know. To speak to this, I reviewed multiple pieces of testing data.
In sum: Test scores have been trending down for over a decade. There are some signs of recovery in math, but not many in reading. Learning declines are not a distinctly U.S. phenomenon and are not even limited to schoolchildren. Researchers are only just beginning to wrap their heads around the causes of this.
Confident claims about what’s going on here are unwarranted, though policymakers can’t wait for perfect evidence to act.
“We should resist the notion of trying to put our finger on the one thing we can change that will solve this problem,” says University of Virginia researcher James Wyckoff, who recently released a paper on declining achievement. “I think it really results from many things in and out of school.”
Here are some key takeaways from my review of the data.
Learning declines have been substantial and pervasive.
Consider one example from the National Assessment of Educational Progress, or NAEP: In 2013, 74% of American eighth graders scored at the basic or above level in math, the highest figure since the test started in 1990. In the most recent round that number fell to 61%, hitting levels last seen in 1996. Scores have fallen in other grades and subjects, too.
Despite a small handful of relative bright spots, these declines have been remarkably widespread. Eighth grade math scores fell in almost every single state during this period; no states saw increases. Although schools that were closed longer during the pandemic tended to experience bigger declines, even those that quickly reopened have been hit hard by learning loss.
This trend started before the pandemic.
Test scores, particularly in math, had generally been marching upwards for a few decades until about 2013. Then a period of stagnation and decline hit. The aftershocks of the Great Recession on families and school budgets may have been an initial cause. Yet even by 2019 there was still no sign of recovery. Then the bottom fell out further after the pandemic.
Two groups have been hit hardest: low performers and girls.
On a wide variety of tests, starting before the pandemic, the gap between the lowest- and highest-performing students has grown. That’s not because high performers have surged ahead but because low performers have fallen further behind.
More recently, since the pandemic, girls’ scores have tended to fall more sharply than boys’.
Some good news: Math scores are starting to trend up again.
Every state with consistent testing data shows that more students are reaching proficiency in math now compared to 2021. Math results have also ticked up on the NWEA exam and on the fourth grade (but not eighth grade) NAEP. Still, most data indicates that these scores have not bounced back to pre-pandemic levels.
There’s been inconsistent recovery at best in reading.
Reading and math results have followed curiously different trajectories. On the most recent NAEP, reading scores actually fell even further. On state exams, reading achievement has been all over the map. Pennsylvania, for instance, has had solid recovery in math, but reading scores have kept sliding downward.
The U.S. is hardly alone in its achievement woes.
Many other countries are grappling with falling test scores too. This has shown up on an exam of 15-year-olds known as the PISA, as well as on the TIMSS, a math and science test of fourth and eighth graders. Relative to the rest of the world, the U.S. trends look a bit worse on TIMSS, but a bit better on PISA.
The U.S. is unusual in its sharply growing gap between highest- and lowest-performing students.
Also: Test scores may be lower among adults and very young children
Some data indicates that children who are just entering school are doing so with lower levels of readiness in reading and math. Another study of adult skills showed drops across the age distribution between 2017 and 2023 in literacy and numeracy.
This adds a new wrinkle. “Factors outside of school might play a considerable role” in learning declines, writes Nat Malkus of the American Enterprise Institute in a report from last year.
How concerned should we be? Pretty concerned!
When children know more, as measured on tests, they tend to lead more productive lives. Countries with higher test scores tend to see stronger economic growth. These scores are incomplete measures of students’ skills, but they do matter.
Test scores are not in entirely uncharted territory, though. A long-running test of 13-year-olds shows that math scores in 2023 were at the lowest point in recent decades but remain comparable to scores from the ‘80s and early ‘90s and higher than those in the ‘70s. Reading scores have dipped to levels last seen in the early 2000s.
So what explains all this? Researchers aren’t quite sure.
Two detailed analyses, by Wyckoff and Malkus, have tried to parse what is driving these trends. Neither concluded with definitive answers. “There is remarkably little understanding of the nature of either the sustained achievement gains prior to 2013 or the subsequent losses thereafter,” writes Wyckoff in his paper, titled “Puzzling Over Declining Academic Achievement.”
That said, it’s very likely that the pandemic and its associated disruptions to life in and out of school played a significant role. Another theory is that easing off school accountability pressure — which research found drove learning gains in the early 2000s — has contributed to recent score declines.
Perhaps the leading hypothesis is the proliferation of phones and screens, although Wyckoff notes that “direct causal evidence” on this question “is limited.” That’s beginning to change.
One recent study linked school phone restrictions to better test scores.
These learning challenges are not particular to American schools and may not even be largely caused by changes within schools. Yet they remain a challenge that schools and educators must confront.
I expect to be writing about potential ideas and solutions for these challenges throughout this year. Feel free to send yours to mbarnum@chalkbeat.org.
