School turnarounds burst into the news last week when the Department of Education released some preliminary “snapshot” data on the School Improvement Grants or “SIG” program. Punchline: Surprise, mixed results!
Skeptics immediately pounced. My colleague Andy Smarick, a former state official himself and longtime turnaround critic, was all over the Department right out of the gate. He’s right, the results are not a ringing endorsement (see for yourself here pdf)) and the Department didn’t do themselves any favors releasing the data the way they did (on the day and in the way you dump bad news). Andy favors more dramatic interventions in school districts, in some cases I do as well, but while the record of new school creation compares favorably to the record for turnaround initiatives it’s far from a surefire fix. On charter schools in particular, plenty of places have managed to make a mess out of chartering. Just because districts are hapless in some places it’s not axiomatic that other solutions will be better.
So we should stipulate that fixing low-performing schools is enormously hard work and reasonable people can consequently look at the preliminary results the department released and see a cup half full or one half empty. There is also other emerging data on the SIG program, Stanford’s Thomas Dee found the natural experiment embedded in the design of the program and exploited that for a study. And Education Sector, Center for Education Policy and others have been tracking the where and how of SIG spending.
Unfortunately our political debate tends to lurch from one “finding” to the next in isolation. And it tends to be defined by the extremes of “this can’t work” to the opposite certainty of “we know what to do here.” The reality probably lies, as it often does, between those poles. Turnarounds have a spotty record, it’s hard to do, most of the politically acceptable strategies don’t work, but it can work and we need bolder supply side strategies to in order to create new higher performing schools. A mouthful? Sure, but this a complicated social policy problem we’re talking about here.
But even more than context, we should be asking which of the four basic approaches under SIG led to the biggest gains or biggest lack of gains, especially if any one approach did disproportionately. Dees data points to more ambitious turnaround strategies. States did different things with this money and it supported different interventions. SIG is not a singular intervention, it’s funding stream. Analytically, that’s where the action is.
It’s also where the reaction to SIG has eerie (and unflattering) echoes of Reading First. A quick rush to judgment about a funding stream. In the case of Reading First it cost students – who needed support the most – a $1 billion reading program. Let’s hope history doesn’t repeat itself here, where the right and the left has reasons to dislike turnarounds. The real answers about what “works” or doesn’t and the lessons lie underneath the overall statistics released last week.
At Bellwether we work with people doing turnaround work, in particular the school turnaround program at the University of Virginia. And we’ve helped recruit talent to some of the most leading-edge turnaround efforts. The results of even the best turnaround initiatives are mixed, too, but caution against throwing the entire idea of turnarounds out.
For my part, I’m increasingly convinced that we’re thinking about turnarounds the wrong way. Many of the interventions states and cities are trying are weak – and they’re trying them on persistently low-performing schools. It seems to me the richest vein for lighter-touch turnaround work, which the field knows how to do, is mediocre schools, not persistently failing ones. In those places you have a core of good teachers to build around, solid support and training can make a difference, and a new leader can lead change.
The lowest performing schools need something more dramatic. That’s where the hardest work is, both substantively and politically. It’s where a strategy of providing new schools can make a big difference, too. But as a field we seem to have trouble talking about turnarounds with that sort of granularity. Probably because no one wants to acknowledge that the other “side” is at least partially right and there are no easy answers here.