Return of the Links: 9/19/25

Woke Charter Schools, SCOTUS Ending Public Schools?, Tutoring Hasn't Worked, Student Perspective on AI in Schools, Handling the Latest Thing

Author

James Shanahan
September 19, 2025

Welcome to Scholastic Alchemy! I’m James and I write mostly about education. I find it fascinating and at the same time maddening. Scholastic Alchemy is my attempt to make sense of and explain the perpetual oddities around education, as well as to share my thoughts on related topics. On Wednesdays I post a long-ish dive into a topic of my choosing. On Fridays I post some links I’ve encountered that week and some commentary about what I’m sharing. Scholastic Alchemy will remain free for the foreseeable future but if you like my work and want to support me, please consider a paid subscription. If you have objections to Substack as a platform, I maintain a parallel version using BeeHiiv and you can subscribe there.

Charter Schools are Woke Now

In yet another sign that the old “treaty” around school reform is well and truly dead, the Heritage Foundation has decided that charter schools are woke now. In a report they released last June (I only just came across it), the Heritage Foundation outlines how charter schools tend to have a high “wokeness score” a measure they created based on keywords written in their student handbooks.

Heritage attributes the wokeness to the role of charter authorizers. Apparently the more regulatory oversight authorizers receive, the more woke the charter schools are.

The Backgrounder [that’s the group within Heritage writing this report] observed that charter schools “in states with more government regulation and oversight are more woke.” Arizona and Florida, which have charter school sectors with robust competition, had the lowest wokeness scores. Meanwhile, even red states like Alabama, Arkansas, Georgia, and Kansas, which have more regulated charter school sectors, had considerably higher wokeness scores.

Why do charter authorizers go along with this so-called ideological dogma? They’re paid by wealthy philanthropies.

The National Association of Charter School Authorizers is a kingmaker within the charter school world. With generous support from the Gates Foundation, Walton Foundation, and millions appropriated through federal Charter Schools Program (CSP) grants, the organization “sets the ground rules” for how authorizers (the entities entrusted with deciding when new charters should open or existing ones should close) should determine which petitions for new charter schools are approved and when charters should be revoked (i.e., when charters should be shut down).

NACSA wields a heavy hand in setting expectations and agendas for authorizer behavior. For example, 10 years ago they endeavored to “close a thousand of the lowest performing charters,” and they continue to advocate for sanctioning authorizers who “do not meet professional standards” set by NACSA.

They argue that the Walton Family Foundation, that’s the Waltons of Wal-Mart fame, have “a left-wing agenda” and that the Gates Foundation “has been pushing critical race theory and other left-wing ideas in education.” It’s honestly a wild read given how much people who actually have left-wing agendas have bashed WFF and Gates for literal decades. Yet, the Heritage Foundation more or less set up the blueprint that the Trump Administration is using to run the country and we’ve seen about half of their reforms or proposals implemented thus far. Their perspectives, however misaligned with reality they may be, are worth taking seriously.

Anyway, what’s their solution to all of this?

  • States should adopt and expand K–12 education savings accounts to foster greater parental choice and competition.

That is, states should do vouchers and let parents “vote” with their (taxpayer subsidized) wallets and their children’s feet. I’m skeptical that this is any real kind of parental choice or oversight and certainly seems only to empower the parents of kids who already attend private schools.

There you have it. The last vestiges of the old school reform movement, the movement that pushed for school choice, charter schools, and vouchers are now on the outside. Conservatives have expertly played them for fools, using their money and influence to break down traditional public schools and sew distrust between communities and their schools so that they can come to power and implement a version of school choice that funnels taxpayer dollars into religious organizations and corrupt pockets.

SCOTUS Ending Public Schools?

Speaking of Heritage Foundation initiatives, historian Johann Neem is worried that SCOTUS may end public schooling as we know it. But what does that mean exactly? In an interview with Jennifer Berkshire, he lays out a few ways he sees this playing out.

We know the court has been bolstering the rights of free exercise over the past few years and has really tried to protect what they consider religious freedom from state overreach. Well, their recent decision in a case over whether parents have the right to opt-out of particular teaching or material on religious grounds continues this trend. The court here said ‘yes they do,’ on the grounds that mere exposure to material that violates a parent's religious faith is a burden on the free exercise of religion. This could end up being an existential issue for public schools because, as Sonia Sotomayor pointed out in her dissent, the very nature of school is that kids are constantly being exposed to messages that could, in theory, conflict with their parent’s religious beliefs. As she wrote, “to presume public schools must be free of all such exposure is to presume public schools out of existence.”

…the case [MAHMOUD v TAYLOR] is not limited to that, which is why it's so scary. The court is arguing that a fundamental constitutional right is at issue here, in this case, the free exercise of religion and parents' rights to raise their kids according to their faith. So we’ve ended up in a situation in which even exposure to ideas that go against parents’ faith could be unconstitutional.

What the court is telling us is that parents, not citizens, should determine the full content of their children's education. It's not that parents have no rights or no claims. I'm a parent. It's that the court suggests that the community, the democratic civic community, has no claim over the socialization of the next generation. And so I think that's a real problem. It basically privatizes the public. It privatizes not just the common schools, it's actually conceptually privatizing the commons themselves by insisting that there are no commons. I think that presumption, that we don’t have a commons that we engage in collectively, is a really dangerous way to approach democratic politics.

If you read my post about the Harper’s Magazine article detailing Arizona’s school choice system (and how people’s hatred of standardization moved them to support right wing approaches to education) then a lot of this probably sounds very familiar. At its root, and we see this in the above Heritage report, the only authority in education is the parent and the only criterion by which education should be judged is whether that education comports with parents’ values. That is a completely, radically different conceptualization of the role of schooling in a nation and the responsibility we, as a society, previously had in ensuring some minimal standard of education across the board.

Tutoring Hasn’t Worked

Readers may recall that I’m skeptical of tutoring as an educational intervention. While there’s some solid evidence that tutoring can deliver strong gains in the right circumstances the phrase, right circumstances, turns out to matter a whole lot. The Hechinger Report covers some of the large scale attempts post-pandemic to use tutoring to catch kids up and undo learning loss (another concept I find dubious).

Even with ample money, schools immediately reported problems in ramping up high-quality tutoring for so many students. In 2024, researchers documented either tiny or no academic benefits from large-scale tutoring efforts in Nashville, Tennessee, and Washington, D.C.

New evidence from the 2023-24 school year reinforces those results. Researchers are rigorously studying large-scale tutoring efforts around the nation and testing whether effective tutoring can be done more cheaply. A dozen researchers studied more than 20,000 students in Miami; Chicago; Atlanta; Winston-Salem and Greensboro, North Carolina; Greenville, South Carolina; schools throughout New Mexico, and a California charter school network. This was also a randomized controlled study in which 9,000 students were randomly assigned to get tutoring and compared with 11,000 students who didn’t get that extra help.

Their preliminary results were “sobering,” according to a June report by the University of Chicago Education Lab and MDRC, a research organization.

The lackluster results are important but I really want to call your attention to something else they mention as a potential reason tutoring fell short.

Monica Bhatt, a researcher at the University of Chicago Education Lab and one of the report’s authors, said schools struggled to set up large tutoring programs. “The problem is the logistics of getting it delivered,” said Bhatt. Effective high-dosage tutoring involves big changes to bell schedules and classroom space, along with the challenge of hiring and training tutors. Educators need to make it a priority for it to happen, Bhatt said.

Some of the earlier, pre-pandemic tutoring studies involved large numbers of students, too, but those tutoring programs were carefully designed and implemented, often with researchers involved. In most cases, they were ideal setups. There was much greater variability in the quality of post-pandemic programs.

and

In the pre-pandemic research, students in the “business as usual” control group often received no extra help at all, making the difference between tutoring and no tutoring far more stark. After the pandemic, students — tutored and non-tutored alike — had extra math and reading periods, sometimes called “labs” for review and practice work. More than three-quarters of the 20,000 students in this June analysis had access to computer-assisted instruction in math or reading, possibly muting the effects of tutoring.

We have two important confounding factors that, I imagine, are pertinent to any proposed educational intervention that’s supposed to “save us”. First, there’s simply the practical challenges of running a good tutoring program at scale and with reasonable fidelity for tens of thousands of students. If your miracle intervention relies on “carefully designed” conditions then you need to spend more time in schools to recognize that all kinds of random shit upends those conditions. Did the kids in the “carefully designed” study have, say, an active-shooter drill randomly during the study?

But, what I personally find more interesting is in the second part noting that 3/4ths of the students in the analysis had access to computer assisted instruction. You might remember something I wrote about effect sizes when combining tutoring and computer assisted instruction:

CAI, which is perennially in vogue, also underperformed 6 other interventions being studied. In fact, the three best intervention categories all included coverage of the content. I want to suggest that this is, actually, quite crucial to understanding when tutoring is successful and when it is not. If all of your tutoring is on test taking strategies or academic skill building, you are probably not going to see the gains that you would from covering the academic content of the course. Likewise, just because you use technology it doesn’t mean you’re going to get better results. Both of these are important given the accelerating combination of tutoring and AI tech.

One way to think about this is to say that adding the scale of AI tutoring is also adding the diminished effect sizes of computer assisted instruction. Yes, you’ll scale but you’ll also have less impact. You’re combining an intervention with an effect just under 0.4 SD with an intervention with an effect of about 0.05 SD. They aren’t additive. Now, I don’t have all the studies from these analyses on hand to try a proper pooled effect size calculation to estimate the effect size of CAI + Tutoring but my best off-the-cuff calculation is about 0.25 SD. It’s not nothing but it’s also not going to revolutionize schooling.

TOT treatment effects for HDT for students who participated across sites ranged from 0.008 SD to 0.13 SD

the TOT treatment effects of SHDT ranged from 0.034 SD to 0.16 SD, again with varying levels of significance.

TOT means “Treatment On Treated” and it’s a way of checking the effects on the treatment group even though some of the treatment group didn’t actually receive the treatment because of non-compliance. Non-compliance in this case is just that they didn’t use the tutoring for long enough to meet a minimum threshold. HDT is High Dosage Tutoring and SHDT is Sustainable High Dosage Tutoring, which refers to “new, innovative lower-cost tutoring models that we co-designed with our partner districts”. Those models, however, varied a bit. For example, in the Chicago group, 4/8 students would work with a tutor and 4/8 would work on the CAI software and then they’d swap. Other places it might have been just CAI or just a human tutor.

Anyway, it appears that my off-the-cuff estimate of the effect size of CAI + Tutoring at 0.25 SD was too high! The actual outcomes look like they trend closer to the historical average of CAI which is roughly 0.05 SD than to the historical average of tutoring alone at 0.4 SD. Tutoring, whatever the form, is simply not going to turn school systems around. You can add that to the long list of silver bullets that failed to live up to their (poorly thought out, poorly evidenced) promise.

Student Perspectives on AI in School

The Atlantic published an essay by a high schooler at the start of the month that I only just got around to reading. She says AI is “demolishing” her education. It’s a fine essay and makes some common points that we should all take to heart but I want to make one point. If we didn’t have computers and phones and all the other devices in schools, we wouldn’t have AI in schools. If we didn’t have assignments “due by 11:59 p.m., to be submitted online via Google Classroom” but rather assignments to be handed in to the teacher on dead trees, then we wouldn’t be in this situation. The wholesale adoption of EdTech leads directly to the ease of AI cheating. The author even asks for the oral exams, portfolio presentations, and pop-quizzes of old. We should take kids seriously when they tell us things. We should listen when they say that they’ll take shortcuts and use AI instead of learning because they have every incentive to do so. And we should understand that our choices around how we structure our schools — relentlessly adopting every digital fad, for example — also entails adopting the logic and ideologies baked into those structures.

Handling the Latest Thing

I do dabble into politics with my writing. Because I am generally a center-left kinda guy, there’s a lot for me to be dissatisfied with and rant about. Sometimes, though, events are so thoroughly… let’s say… fucked, that I don’t exactly know how to write about them. So, in lieu of adding to the cacophony of perspectives on the murder of Charlie Kirk, I will share three links without commentary.

Thanks for reading!