🛝 Three Things Everyone Knew. The Data Disagrees

Welcome to Playground Post, a bi-weekly newsletter that keeps education innovators ahead of what's next.

This week's reality check: The first real proof of post-COVID academic recovery just landed, and it reached only the youngest kids while 13-year-olds sit exactly where they were in 1971. California's community colleges paid $1.9 million in financial aid to students who don't exist. And a Stanford study found that closing a school, the move every cash-strapped district reaches for, saves almost nothing.

Data Gem

27% of US adults ages 25 to 64, or 41.7 million people, intend to enroll in college or job training within the next two years, according to a new report from CAEL and College APP. Extrapolated across all adults 18 and older, the intent pool tops 65 million.

9-Year-Olds Are Finally Recovering. 13-Year-Olds Haven't Moved

The Nation's Report Card just delivered the first hard evidence that students are climbing out of the pandemic.

But only some of them.

The Long-Term Trends edition of the National Assessment of Educational Progress, the longest-running measure of American student learning, showed the average 9-year-old gained 4 points since 2022.

The average 13-year-old gained nothing.

Scores for that group were flat or falling across nearly every demographic group. Their reading performance is no higher than it was in 1971, the year the test was first given.

The split traces to timing. Today's 9-year-olds were in preschool when COVID hit and largely missed the worst of the closures, while today's 13-year-olds were second and third graders in 2020 and carry the scars of years of disrupted instruction.

What makes the recovery notable is who drove it. Nine-year-olds at the 25th percentile gained 7 points in math and 6 in reading since 2022, and those at the 10th percentile gained even more.

Through the 2010s, only the highest-performing students made gains on NAEP. 

This time, the lowest performers led.

"They were in some of their most formative years of both literacy and numeracy, and it was a seismic event," said Kirsten Baesler, who leads the Office of Elementary and Secondary Education at the Education Department. "It's going to take equally seismic effort to ensure that those students are coming back to where they need to be."

The teenagers offer no such bright spot, and the decline that produced their scores predates the pandemic.

In 2012, 85% of 13-year-olds could clear the math benchmark for solving a one-step word problem. By 2025, only 70% could, and the share reading at benchmark fell from 66% to 58%.

Researchers from Dartmouth, Harvard, and Stanford have labeled the stretch since 2012 a "learning recession," with achievement falling for everyone except the top students. 

Among 13-year-olds, the share absent at least one day in the previous month has climbed from 44% in 2012 to 61% in 2025.

For education innovators, the split is the product spec. The recovery tools built for early grades worked, and the data proves it. The opening is for middle-school catch-up curricula designed for 11- to 13-year-olds rather than repackaged elementary content, reading-engagement platforms built to compete with the phone for a teenager's attention, and diagnostics that separate a student recovering on schedule from one stuck in the cohort that never moved. The market that treated "learning loss" as one undifferentiated problem just learned it is at least two.

California Colleges Paid $1.9 Million to Ghost Students

In the first three months of 2026, California's community colleges handed out more than $1.9 million in financial aid to students who don't exist.

Not students who dropped out, but identities that were never real.

The 116 colleges in the California Community Colleges system have lost $30 million to the scheme since 2024. 

Scammers use stolen or fabricated identities to apply, enroll, and collect federal aid, usually at community colleges where open admissions make entry easy.

Last week the House responded.

The No Aid for Ghost Students Act passed 249 to 172 - it would require the Education Department to screen federal aid applicants for identity fraud, codifying a system the department quietly launched in April.

Under it, applicants flagged as high risk verify themselves through the online FAFSA form, presenting a government ID and taking a picture in a live camera check. Those who can't do it online verify in person at their college.

The department says the early version of these measures already prevented $1 billion in fraud.

"The No Aid for Ghost Students Act makes those protections permanent, helping ensure that federal student aid goes to real students working to further their education, not scammers gaming the system," said Rep. Burgess Owens, the Utah Republican who introduced it.

But the fix creates a problem of its own. Real students get caught in the net.

Rep. Bobby Scott of Virginia warned that applicants wrongly flagged and blocked from their funds "could face hardships obtaining housing, transportation, or even buying books for who knows how long." He also argued the system is too new to evaluate, noting it "is still being tested, and Congress has not yet seen meaningful evidence about its effectiveness."

For education innovators, ghost-student fraud is a verification problem the largest community college systems are now legally required to solve.The openings are identity-proofing tools built for FAFSA's live-camera workflow, fraud-risk scoring that flags fabricated applicants without burying real ones in false positives.

Closing a School Was Supposed to Save Money. It Doesn't

When a district runs out of money, the playbook says close a school. Fewer buildings, lower costs: less to heat, light, and maintain.

It is the first move districts reach for when enrollment falls and budgets tighten.

A new Stanford study says it barely works.

Francis Pearman, an assistant professor at Stanford, examined every school closure in California between 2011 and 2019. Closures cut per-pupil spending by about $440. They also cut per-pupil revenue by about $440.

The district breaks even.

The reason is enrollment. When a school closes, families don't all stay in the district, and the study found closures cost districts an average of 287 students over the following five years. 

In California, that works out to roughly $2.4 million a year in lost funding.

The savings, meanwhile, never fully arrive. Districts usually reassign teachers rather than cut them, so staffing, the biggest line in any budget, stays flat. The empty building still needs upkeep and money to prep it for sale.

Earlier research found districts typically save less than 5% of their budget by closing a school.

The findings held across districts of every size, location, and demographic makeup. The break-even result wasn't an exception. It was the pattern.

For education innovators, the closure-savings myth is a forecasting gap, with districts making nine-figure decisions on an assumption the data just contradicted. The openings are closure-modeling tools that project the enrollment exodus rather than just the facilities savings, scenario planners that show a school board the real five-year net, and enrollment-retention analytics that flag which families are most likely to leave before the vote rather than after.

⚡️More Quick Hits

This week in education:

• Early math mastery predicts who passes Algebra I, a study of 1.7 million Texas students finds — AIR and CALDER researchers traced Algebra I success, and the racial and income gaps in it, back to foundational skills in early elementary grades, moving the intervention window years earlier than most secondary-school policies target

• A CUNY peer-mentor program cut summer melt to 7.7%, down from 14.4% — Of nearly 40,000 NYC public school graduates who planned to enroll last fall, 92.3% matriculated through the nation's largest near-peer advising program, which staffs the gap between admission and the first day of class

• House moves to let graduate nursing students borrow $200K instead of $100K — The Education Department's rules excluded grad nursing from the "professional degree" category that permits double the federal borrowing, a distinction that has already drawn at least three lawsuits, including from more than two dozen states and nursing organizations

• Schools with well-prepared principals retain teachers at 89%, versus 78% elsewhere — A synthesis of two decades of research ties leadership quality to teacher retention, chronic absenteeism, and student achievement that can keep declining for up to five years after a principal leaves

To stay up-to-date on all things education innovation, visit us at playgroundpost.com.