Kevin Drum expresses frustration with the very mixed evidence on charter schools:
I’ve been modestly favorable towards charter schools for a while, and I still think they’re worth trying. It might take more than a few years to get the formula right, after all, and most of the research suggests only that charters don’t outperform public schools, not that they’re actively worse. (The Stanford study showed mixed results, with better results for charters in grade school and middle school but worse results in high school.) Still, time is running out. If charters can’t start demonstrating systematically better results soon, the experiment is going to run aground.
I think that’s really the wrong way to think about it. If you look at the charter school system in a typical jurisdiction with low-performing public schools, what you usually have is an oversubscribed charter sector combined with a statutory cap on the number of charters that are allowed to open. Some of these charter schools may perform well and others perform poorly. And since the public schools are performing poorly, even the low-performing charters have an okay time attracting students. Meanwhile, the higher-performing charters are helping the kids who attend them, but it’s necessarily a small number of people.
The solution to this isn’t to say that “the charter school experiment” has “run aground.” The solution is to scrap the existing cap policies and replace it with something more like smart caps that are actually focused on school quality. I don’t think it should surprise anyone that charter schools, as currently administered, perform about the same as public schools. But it’s the very averageness of currently existing charters that provides the opportunity for improvement. On average the charters are about the same as public schools, but there’s a range of outcomes within the charter sector. We need to get more aggressive about shutting down the low-performing charters, more aggressive about allowing successful charters to expand or replicate, and committed to always permitting space for people to try something new.
The “something news” that people try probably won’t be any better, on average, than existing public schools. They might even be worse! But then you shut down the models that don’t work and let the models that do work replicate. There’s no “charter magic” that makes schools good, but the greater openness and flexibility of the charter sector lets us experiment and discover which things work. What we need to do is take that to step two where we act on the basis of that knowledge.