I’m not that surprised that the 3rd year evaluation of the D.C. voucher program (pdf) found modestly positive results, that’s in keeping with other research on the issue. I am surprised that Andy Smarick thinks evidence will have anything to do with the debate:
Of course, there will be lots and lots written and said about this over the next several days, and of course this all will have a bearing on the reauthrorization [sic] of the program…
Right! I’d forgotten how empirical the debate on the Hill was…
For voucher opponents the program is like that scene in “Saving Private Ryan” where the Germans keep shooting the runner to make sure the message dies with him. As long as the voucher program lives it carries a message, they must stop that. Proponents feel the same intensity, just in the opposite direction.
That’s why, bottom line, this seems unlikely to change any minds. Previous D.C. voucher posts here.
Update: WaPo ed board turns up the heat.
I’m not suprised either, as previous random assignment voucher studies found precisely the same thing: year by year gains that accumulate over time until they reach the level of statistical significance.
In the year two evaluation, if you only looked compared kids using vouchers to those who lost the lottery, you already found a significant effect on reading. In the year three evaluation, the kids were scoring high enough to clear the intention to treat bar.
The department’s press release claimed that there was no impact on math, but it is much more accurate to say that there hasn’t been a statistically significant impact on math yet.
President Obama told the U.S. Hispanic Chamber of Commerce:
“Secretary Duncan will use only one test when deciding what ideas to
support with your precious tax dollars: It’s not whether an idea is liberal or conservative, but whether it works.”
Get ready to rumble indeed.