SIG: A Disappointing But Completely Predictable Reaction From Smarick
Researcher Bryan Hassel has written a bracing (for policy wonks) response to yesterday's "SIG-failed-I-told-you-so" post from former New Jersey state education official Andy Smarick (The disappointing but completely predictable results from SIG).
In his rebuttal, Hassel questions Smarick's contention that SIG has failed and shreds Smarick's notion that starting new schools is a viable way to go:
"There’s no evidence that new school creation is demonstrably better as an overall strategy than turnarounds... To replace the 5,000 worst schools, we’d need 10,000 high-quality new schools b/c they tend to be smaller."
Read the full post below.
As usual, Andy is half-right on this stuff. He’s half right b/c of course far too many SIG schools have followed the pattern set in past waves of this work, under names like Restructuring. They pursued incremental changes like providing more PD for their teachers, bringing in new curricular programs, and the like, none of which is very likely to lead to transformative change in a dysfunctional school. Far too few SIG schools have brought in carefully selected leaders or organizations with the capacity to lead turnarounds, and given them room to do what’s needed.
Andy is also right that we shouldn’t be surprised by this, since the political deck is stacked in most school districts against doing something more fundamental. In theory states could use SIG to tip the balance, and some have more than others, but in most places that won’t tip enough.
But Andy’s only half right because of his conclusion that therefore we should stop trying to fix failing schools and put all our eggs in the basket of new school creation. This has never made sense to me for 2 reasons. First, there’s no evidence that new school creation is demonstrably better as an overall strategy than turnarounds. Andy, if we divided charter schools into three groups, those that were “double digits” better than comparable district schools, those that were single digits better and those that were worse, where would the proportions fall? Would they look a lot better than the SIG numbers? If so, show me.
Second, related, the supply of high quality new school operators is ridiculously small relative to the need. To replace the 5,000 worst schools, we’d need 10,000 high-quality new schools b/c they tend to be smaller. We haven’t even created 10,000 charter schools in total in 20 years, much less 10,000 great ones. I’m sure we could create them faster but I don’t see your path to solving this problem entirely with new school creation. How many great new schools can we create in the next 5 years in your wildest dreams? 500? 1000?
All of this isn’t to say we shouldn’t be dramatically increasing the use of new starts as a way to address chronic failure, something I’ve been advocating since my 2003 paper with Lucy Steiner called Starting Fresh. But the reality with both new starts and turnarounds, they can be done well or badly. Doing another round of turnaround attempts that’s just warmed over school improvement is a bad idea, but so is doing another round of new starts that replicates the charter performance curve. Wouldn’t it make more sense to zero in on the best of both and do a lot more of both strategies?