Matt Yglesias asks this question about the series of revelations that jumps in test scores that occurred under the supervision of Rod Paige, Michelle Rhee, Beverly Hall, and other big-name education 'reformers' were based on cheating.
The weird thing is, we both reach the same answer: nothing.
The unsurprising thing, I guess, is that our meanings are 180° apart.
Matt's implicit assumption is that high-stakes testing regimes are a Good Thing, and if there were a few bad apples, that's insufficient evidence to justify upsetting the testing applecart.
My question is, where's the evidence these high-stakes testing regimes were ever shown to be a Good Thing to begin with? We've known for a decade that the "Texas Miracle" that got Rod Paige the Secretary of Education job under GWB was all about cooking the books. If the other major 'successes' touted by the 'reformers' - Rhee, Hall, and whoever else - are also based on cheating, then you have nothing. You have no reason to prefer this program over the status quo ante. None.
So Matt, that's what cheating proves: nothing. Anytime you conduct an experiment and lie about the results, the experiment proves nothing.
Now maybe someone, somewhere has conducted an experiment in high-stakes testing and it's produced evidence that high-stakes testing works. But if a whole bunch of well-known experimenters trying to prove the same result have all claimed success, and have all been later shown to have achieved their results by cheating, then you really might start wondering why they're having such a hard time demonstrating success without cheating.
Maybe the claimed result isn't even true. And whether it is or isn't, there's no reason to believe it's true, there's no reason to act in ways that assume its truth.
And anyone interested in what's going on with education 'reform' should be reading Kay at Balloon Juice.