But while I have no problem with the idea that there should be consequences for Beverly Hall or Michelle Rhee or any other school chancellor who presides over cheating, I’m genuinely puzzled by what anti-reform people think these cheating scandals prove.
Well, first of all, the Rhee scandal seems directly relevant to the claims of the “reform” movement given that the alleged gains in student achievement under her tenure were often cited in defense of “reform,” and Rhee herself remains a highly influential figure in “reform” circles. The fact that the alleged gains in achievement under her tenure were almost certainly the product of cheating seems highly relevant to whether her “reforms” are effective to me.
This also seems like missing the point:
Now if you wanted to say that these cheating scandals prove that we’re never going to come up with a workable control system for organizations as large as big city public school systems and so we need to move to an all-charter system, I’d say that’s an idea I’m sympathetic to.
Although perhaps some people have made this argument, I don’t think it’s the central issue. Since it’s possible for national tests to have high degree of integrity, I’m not sure why it’s impossible for local ones to. The point of the Rhee scandal is that the lack of testing oversight wasn’t just incidental incompetence. Rhee 1)used high-stakes tests as the sole criteria* for bonuses for ongoing employment for many teachers and administrators, and perhaps even more importantly 2)rather than using high-stakes testing to measure progress towards incremental improvements over time used them to demand ludicrously implausible immediate improvements in student performance. So of course there was going to be widespread cheating. Not only did educators and administrators need it to save their jobs, Rhee needed the cheating because without it she couldn’t have claimed the phony massive immediate improvements that made her a star.
There’s no reason that it’s impossible to have fair, well-monitored testing in a large school system. There is good reason to doubt whether extremely high-stakes tests will be applied with integrity, but since Rheeism is largely based on extremely high-stakes tests meant to show flashy immediate gains in some cases and results that can justify mass firings in other cases, this seems relevant to whether Rheeism is a good idea.
And finally, a classic fallacy of the excluded middle:
What about the fact that some people respond to performance-based systems by cheating should make me think that pure seniority systems are good?
I concede the point: the Rhee cheating scandal cannot prove that there should be no attempt at all to evaluate the performance of teachers beyond showing that they’re competent enough to get tenure. What I do certainly dispute is whether the only alternatives are “evaluating teachers using high-stakes testing as the sole criterion” or using a “pure seniority system.” Using standardized testing as part of a fair, well-constructed comprehensive system of evaluating teachers is a perfectly good idea. But that’s not what Rhee-style “reform” generally consists of.
Undeterred by the release of John Merrow’s report of widespread cheating on her watch, Michelle Rhee traveled to South Carolina to attack teachers. She said they were defenders of the status quo. She said they were protecting their self-interest. She said they ride a “gravy train.”
The average teacher’s salary in SC is $46,306.67.
Rhee is paid $50,000 for lecturing and taking questions for an hour.
*UPDATE: MY writes to note that the current D.C. system, originally implemented by Rhee, does not use test scores as the sole criteria of value. This is my error; I was referring to the fact that administrators had to commit to large test score gains or be fired, but teachers, at least on paper, are not solely evaluated by test scores. It’s hard to tell what weight qualitative measures receive in the overall evaluations, but they’re at least not formally absent. My point was overstated, and I regret the error.
To make the point in the more subtle form I should have in the first place, the legitimate concern with the introduction of high-stakes standardized testing is that the superficially persuasive-looking quantitative data will tend to swamp other, more qualitative standards that require more time, resources, and expertise to evaluate. This certainly seems to be the case with Rhee in practice. But this isn’t inevitable; it’s about the design and the application.