Tuesday, March 29, 2011

Ravitch: Pay-for-test-scores "never works"

When Diane Ravitch takes a position, she doesn't disappoint. In fact, she's a primary example of why lawmakers want badly to get rid of veteran educators: The longer you remain in your career, the more you learn, the more effective you can become, and the more quickly you can summon to mind the resources and references that support the work you do. Ravitch is like that -- a walking encyclopedia of research, studies, conclusions and findings -- and when she turns her focus to such a scheme as pay-for-test-scores, she brings 40 years' worth of homework to bear.

That's the topic of her current blog post: "Thoughts on the Failure of Merit Pay," and I strongly encourage readers to read her post at her website, because she hyperlinks many, many of the studies that she mentions there.

"It is curious that teachers vigorously oppose merit pay," she writes, "even though they are the ones who are supposed to reap the rewards. What do they know? They know that merit pay undermines collaboration and teamwork. They know that it corrupts the culture of the school."

But corporate reformers think they know best, so they continue to push for a reward system that will give bonuses for "effective" teachers (those whose students get higher test scores) and, thus, magically, make teaching attractive to the theoretical "best and brightest," those graduates of Harvard, Princeton, Yale, and Stanford who would stay in teaching if only they could compete every year for an extra $5,000 or so.

Then, the data cascade: Ravitch uncorks history.

Merit pay has been tried again and again since the 1920s. Sometimes scores go up, sometimes they don't, but the programs never seem to make much difference and eventually disappear.

The most rigorous trial of merit pay was conducted recently in Nashville by the National Center on Performance Incentives. It offered an extraordinary bonus of $15,000 to teachers if they could get higher scores from their students. Over a three-year period, there was no difference between the scores obtained by the treatment group or the control group. The bonus didn't matter.

Roland Fryer of Harvard University just released his study of New York City's much-touted school-wide merit-pay program. Fryer says it made no difference in terms of student outcomes and actually depressed performance in some schools and for some groups of students.

So the range of effects from these schemes in the past has been, on one end, the bonus didn't matter because scores wouldn't increase anyway, to the other end, where student performance suffered.

Not a ringing endorsement of Governor Nikki Haley's and Superintendent Mick Zais's scheme to tie teacher salaries to test scores.

Yes, yes, I know: South Carolina will be different. Our state has always been different. What doesn't work for any other state, anywhere, will work here because we'll make it work by sheer force of will and because our lawmakers say so. Until, that is, it buckles under the weight of its own stupidity and fails, at which point it'll be blamed on other influences, like Yankee intervention or Northern agitators or lack of citizen commitment, or the fundamental inability of certain groups of people to learn. Institutionally, South Carolina has never been wrong.

Well, pay-for-test-scores has been proven wrong when it's been tried elsewhere, as Ravitch explains. And here's a reference that should tickle the ribs of the state's corporate elite:

Andrea Gabor wrote an opinion piece recently for Education Week describes what she learned from the work of W. Edwards Deming, the eminent business consultant. I learned a lot by reading her book The Man Who Discovered Quality, about Deming's philosophy. Chapter 9 summarizes Deming's strong opposition to merit ranking and merit pay. Bottom line: It is bad for corporations. It gets everyone thinking about what is good for himself or herself and leads to forgetting about the goals of the organization. It incentivizes short-term thinking and discourages long-term thinking.

And a love letter to the Libertarian free marketeers among us:

Dan Ariely's Predictably Irrational explains why money is not as good a motivator as a sense of purpose. Ariely is an economist of human behavior at the Massachusetts Institute of Technology, and he has demonstrated again and again that people will work harder for idealistic reasons than for a promise of money. He warns of the danger of shifting education from "social norms" to "market norms." This is precisely what the corporate reformers want to do.

And on the opposite end of the academic spectrum, a note for the human behaviorists in the room:

And then there is Daniel Pink, who has written Drive and also created a delightful YouTube video, to reinforce the view that a sense of purpose, the desire to make a difference, is a better motivator than cash on the table.

Again, I highly recommend going to Ravitch's column for yourself and clicking through the links she posts there to read the source material.

THEN, just for kicks, why not send the link to this post to YOUR member of the Legislature? (A) It's an easy way to communicate your thoughts on pay-for-test-scores (B) without having to talk to him or her yourself and (C) with plenty of source material available to prove your point with simplicity and elegance.

If he or she isn't a reader and needs you to break down the point for them, tell them Diane Ravitch says:

Merit pay is the idea that never works and never dies.

No comments:

Post a Comment