Wednesday, February 17, 2010


So now in mid-February 2010 the school board in Houston has approved a policy to allow the district to fire teachers whose students fail to show adequate growth on the state's standardized test. Meanwhile, the Czar of New York City -- sorry, make that hizzoner Mr. Bloomberg -- and his grand vizier -- I mean Schools Chancellor Joel Klein -- have decided that, for those teachers who are in the bottom quartile of performance (determined how exactly?), student test scores will "inform tenure decisions." Both ideas have met some resistance from teachers, though not as much as you'd expect from their unions.

Now, to take the Houston case, tracking student growth seems to make sense. But the valued-added technology is not yet as good as many of its proponents would like it to be. In fact, value-added testing or the use of growth models may never be good enough to allow us to say with confidence how much of a student's growth is caused by anything the teacher did or didn't do. (Forget for a moment whether the test can measure growth that really matters.) Pretending that the problems have been resolved won't make them go away, no matter how many school boards, state legislatures, or even U.S. congresses vote that it should be so.

At the moment, both test scores and other kinds of measures (yes, there are some!) ought to be made available in a timely way for formative purposes. But certainly they're not ready for the prime time of summative assessment, with the highest stakes possible attached. Student failure rates reflect many influences, only one of which is the performance of the teachers. Student failure rates are also intimately connected to poverty levels, extend across generations, and will only be exacerbated by ratcheting up the stakes on yet another quantitative indicator.

Already home to numerous scandals with regard to data manipulation and outright cheating, the state of Texas and the city of Houston -- and soon many other large cities -- will be subjected to another round of policy makers' chanting in self-congratulatory ignorance: "Campbell's Law? We don't need no stinking Campbell's Law!" But the distortions will follow as surely as the night follows the day.

To the AFT, allegedly the protector of the interests of teachers in both Houston and New York City, there's only one sensible message: Just say no! Sometimes you have to put your money where your mouth is, and bartering your purpose for a "seat at the table," which has been the ineffectual modus operandi of the NEA for a generation, is unworthy of our teachers or their representative organizations.

Moreover, as more states twist themselves into contorted shapes in order to qualify for Race to the Top funds, we see a range of serious abuses of teachers and the children they are responsible for. But rather than continue to rant on my own, let me share with you what my late friend Gerald Bracey said on the matter as recently as September 2009, just six weeks or so before he died. Here's the text of an e-mail from Jerry, which I'm glad I retrieved from my Sent Folder:


The President of the United States and his Secretary of Education are violating one of the most fundamental principles concerning test use: Tests should be used only for the purpose for which they were developed. If they are to be used for some other purpose, then careful attention must be paid to whether or not this purpose is appropriate. This position was developed jointly by the American Educational Research Association, the American Psychological Association, and the National Council on Measurement in Education in their document "The Standards for Educational and Psychological Testing." The President and his Secretary want to use existing tests, willy-nilly to evaluate teachers. They should both be ashamed. The President should be chastised and Secretary Duncan should be fired on the simple grounds of incompetence. From the American Psychological Association’s “Appropriate Use of High-Stakes Testing in Our Nation’s Schools:” “It is important to remember that no test is valid for all purposes. Indeed, tests vary in their intended uses and in their ability to provide meaningful assessments of student learning.”

It is not the case that a test is a test is a test. Research into the use of existing tests to evaluate teachers has called that use into question. (I summarized other people’s research on these issues in my May and December 2004 Research columns in Phi Delta Kappan.) More recent studies, summarized in the July 15 issue of Education Week by Debra Viadero, cast further doubt on the whole enterprise. States are rolling over and playing dead on this issue because a) they are desperate for money and b) it is unlikely that people like Bloomberg or the Governator—or Duncan-- have a clue about the abuse they are permitting and advocating. Duncan’s enthusiastic championing of a “reform” that has been shown not to work very well—charter schools—can only be taken as an instrument for union busting. If the NEA and AFT won’t stand up to this abuse of testing, they deserve to be busted.


But the madness continues unabated. I don't fancy myself to be a prophet, but I will go out a limb and I say that this, too, shall fail.

Those who want to follow some really thoughtful exchanges on this an other topics related to the current silliness ought to visit the Bridging Differences blog, hosted at the New York Times website. There, Diane Ravitch and Deborah Meier find that their differences are a lot smaller than many folks would have predicted. Of course, when the subject is the errant nonsense being foisted on educators by those in Washington and the nation's state capitals, it's fairly easy for two sensible people to find common ground -- and to keep their feet firmly planted thereon.

No comments:

Post a Comment