Wednesday, February 14, 2007

Is The Teacher Teaching?

WASHINGTON (AP) -- An updated No Child Left Behind law should track the progress of teachers as well as students, a special commission said Tuesday.

The private commission said schools should be required to measure how well teachers are doing at raising student test scores -- one of 75 recommendations in a report meant to guide Congress as it prepares to rewrite the 5-year-old law this year. The idea came under immediate attack from the nation's largest teachers union.

Teachers should be evaluated annually based on progress in the test scores of their students, the panel said. Reviews by colleagues or school principals also would be part of the equation for determining teacher quality.

As big a supporter of NCLB that I am, this I have issues with.

One of the chicken-crap complaints about proposals like this is that it would force teachers to "teach to the test". But those are just words without real meaning. How *does* one teach to the test, unless one has the test and teaches how to solve the problems on it? Don't give teachers "the test" in advance, and you don't have to worry about teaching to it. I teach to the state standards, which are what (I assume!) will be tested--but that's entirely different from teaching to the test. Think of it like driver's training--it's not teaching to the test to teach students what's in the DMV handbook; in fact, that's called good teaching! Teaching to the test would be covering specific questions on the test (which is sometimes done, as I recall from 1981!).

So what are the real problems with this type of suggestion? In a perfect, regimented, top-driven world, absolutely nothing. In our country, plenty.

I'm not going to go into detail here--it's getting too close to my bed time. Let's just lay out some of the obvious problems.

1. Each state uses different standards and gets to choose its own tests. Wouldn't we be comparing apples and oranges?
2. I support federal accountability, and agree that states could forego federal education dollars if the strings become too burdensome. Since this proposal would require a change to California's ed code, which says that teachers cannot be evaluated based on student performance, would that be too burdensome?
3. Are teachers the only influence on whether or not students learn; are they the most important or most powerful influence? If so, fine. If not, does this proposal accomplish anything?
4. How much control do we want to give the feds in this non-enumerated power?
5. How do you keep this kind of power from being abused by administrators?
6. How do you deal with naturally "slow" kids, and the teachers who teach them?
7. Do we expect all teachers to be from Lake Wobegon? Is competence to be defined as excellence? Excellence is the standard, but is everyone who is good, but not excellent, incompetent?

Is Tennessee's "value added" metric applicable here? I don't know. But until the above questions are answered, someone's just talking and not thinking through the details--and we all know what's in the details.

Update, 2/16/07: Joanne (see blogroll at left) has more commentary.


Mike said...

Dear Darren:

Just a couple of problems here. How does one teach to the test if one doesn't have the test? Easy. Move to Texas. While our annual TAKS tests are sheathed in a veil of secrecy that puts the CIA to shame, we do know the type of test items (and here I speak of the English tests) that will appear and the criteria that will be used for judging those items. The state also releases older tests from time to time.

And while everything on the tests can be said to adhere to the state standards (they're called TEKS in Texas), one must absolutely teach to the test if students are going to pass. There are indeed very specific tricks that must be used if a student is going to pass. If a student isn't taught very specific ways of writing on the various types of item--most of which do not translate to good writing practices in general--they can easily fail. I've seen some of the finest writers in school, some of the most intelligent kids extant, fail. By the same token, I've seen kids who could barely sign their names apply the tricks well and pass. Fortunately, our English department has correctly figured out the tricks and is quite successful at teaching them, but doing this is not compatible with good, general English teaching within the curriculum. We absolutely must take a month or so prior to the test to drill the kids.

That said, we come to another problem. Our English department commonly sees 95%+ of all kids passing the test. It would be awfully hard for us to improve much on that, particularly considering that from 12-15% of our school population is classified under special education. Under such rules, it's entirely likely that we could find ourselve not making adequate yearly progress/improvement.

But as I've asserted before, all of this is properly a local problem, solved locally, is it not?

Darren said...

What you've described is the fact that Texas has a horrible testing instrument--which is a totally different problem than Washington's wanting to see that it's getting something for the federal dollars spent.

Nic said...

I teach foreign language, none of which are on our state testing (which test reading, writing, and math, in English). I always wonder where exactly I (and the Gym teacher, the Health teacher, the Computer teacher, etc etc etc) fall in these schemes.

Darren said...

Another excellent point, Nic.