Hello from Erica and Lily on Chalkbeat’s national desk. Dueling ballot measures, social media warfare, an incumbent state schools chief ousted — the battle over the future of Arizona’s voucher-like program brought high drama to the desert and exposed fractures among conservatives who all support school choice. That’s our big story this week. 

Keep reading for that, plus great test score mysteries of 2026, the higher education opportunity gap, and a new survey looking at the well-being of LGBTQ youth. 

We’re running a back-to-school membership campaign to invite readers who value Chalkbeat’s reporting to help sustain it. If our journalism helps you understand what’s happening in schools and why it matters, consider becoming a member today so our reporters can do this work all year long. Support Chalkbeat at a level that feels comfortable for you.

A new school year needs strong education reporting. Help Chalkbeat’s reporters keep asking the questions that matter.

Arizona has been a proving ground of school choice for many years. 

The supporters of the state’s Empowerment Scholarship Accounts say they underscore Arizona’s status as an innovation hub, where microschools and a vibrant homeschool community have been allowed to flourish — without the government breathing down their necks. 

To critics, the Grand Canyon State’s overly permissive policies have created an environment where parents can spend money earmarked for education in questionable ways, like trips to SeaWorld, or on even more dubious expenses like diamonds. This group has been pushing to change the rules.

A union-backed ballot measure, the Protect Education Act, sought to beef up oversight of the ESA program. But the Arizona Supreme Court dealt a fatal blow to that proposition Tuesday and effectively kicked it off the ballot.

The oversight debate doesn’t end there, though. Current state Superintendent Tom Horne was ousted in the Republican primary last month by state Treasurer Kimberly Yee.

Yee has been walking a fine line in her remarks about oversight, critiquing the management of the program by Horne while also accusing him of overstepping his authority in denying purchases. Now she faces Democrat Teresa Ruiz, who says she’d create a dashboard displaying every ESA purchase for full spending transparency.

Other states are grappling with these questions. And they’re likely to emerge elsewhere as new programs pop up (howdy, Texas) and expand (lookin’ at you, Tennessee). Not to mention a federal education tax credit on the horizon.

Conservatives battled the Protect Education Act every step of the way. But it also revealed divisions between voucher supporters who don’t want any regulation, and some who might be open to it, even if reluctantly.

State Rep. Nancy Gutierrez, a Democrat, told Chalkbeat that some Republicans do want oversight. Admitting it, however, has proven politically risky. 

“Even behind the scenes, some of my colleagues on the Republican side agree that there should be more accountability and transparency,” she said. “But they won't vote for it.”

Read the full story here.

Affluent students tend to go to colleges that spend more on their education. New research documents how the rich get richer, so to speak, even in their academic opportunities. The most selective schools — whether they’re elite private colleges or flagship public universities — admit the students with the best test scores, who tend to come from wealthier families. And those schools, in turn, devote far more financial resources to their students’ education, widening opportunity gaps that begin in early childhood. This is uncontroversial and often viewed as meritocratic.  

New York City schools officials cheered last year when students showed big gains on state reading tests. Efforts to improve curriculum and instruction to better align with the science of reading must be paying off, right? 

That made it pretty awkward when elementary reading scores plummeted in the city and state, according to results released earlier this month. Is the reading overhaul not all it’s cracked up to be? Did teachers change how they were teaching? Or did something go wrong with the test? 

Education economist Emily Oster questioned the plausibility of the score changes in a post on Substack and strongly suggested there was a technical error. NWEA, the state’s testing vendor, says it hasn’t detected any anomalies. A state education official pushed back on Oster’s theory, saying “speculation is not a substitute for data.” 

To be clear, there’s no hard evidence that there was a testing error. Yet some experts say it’s unlikely such severe swings represent year-over-year changes in student learning. Lots of folks say it’s important to get to the bottom of this, especially as New York City’s teachers union threatens to pull support for the city’s reading reforms. 

The uncertainty highlights something else uncomfortable. We want concrete ways to measure student achievement, but even supposedly objective measures like test scores involve methodological and technical choices that require judgment calls and can introduce errors.

Less mysterious, but more common, is that many states have changed the cut scores they use to determine which students are proficient in recent years. That’s made year-over-year comparisons essentially impossible and cast doubts on claims that more students are learning what they need to know.

Meanwhile, test scores are delayed in New Jersey this fall, meaning teachers and administrators are starting the school year without key data. And in Philadelphia, district leaders say they’re also waiting on the state to release test score results, while the state says the district has had preliminary data since June. 

That’s how many transgender, nonbinary, and gender-expansive youth reported being mistreated so badly they left school entirely, according to surveys conducted by the advocacy groups The Trevor Project and Glisten. 

Their joint report describes significant challenges with safety and belonging at school. More than half of respondents reported attending a school with at least one discriminatory policy. But 97% of respondents described having at least one supportive teacher at school. 

That was cognitive psychologist Ronald T. Kellogg, talking to the New York Times about what students gain from doing their own writing and what they lose when they let AI do that work for them.