PUBLISHED: October 26, 2006 | By BENJAMIN WACHS | Messenger Post Columnist

Ever since I started covering schools in New York state, I've dreaded this time of year.

Every fall, like a nasty cold, the State Education Department releases its various test results from the year before, and it catches. Soon every school, television station and newspaper for 100 miles is coughing out the question: "What do these results mean?"

Every year they come up with some kind of answer, and every year it's dumber than the last. Because. . . the truth is. . .

The test results don't mean anything.

They are impractical, unscientific and a colossal waste of teacher time and taxpayer money. . . but the media haven't figured it out, and school districts don't want to rock the boat. So we all go through this tired old Kabuki theater version of "Dick and Jane fudge the data" until we drop from exhaustion.

This year will be just the same, and here's why you shouldn't participate.

Problem 1: the tests come out too late.

The tests were taken last school year and the results came out when? This month. Long, long, after decisions about the fate of children were made and it's too late to do anything about it.

What if the tests indicate that Johnny needs to be held back a grade? Well, it would have been nice to know. What if the test results say Suzie needs summer school? Tough luck: it's fall now.

By the time the tests come out, students are already where they're going to be for the year, and either doing fine — in which case there's no problem after all — or having trouble in school, which we already know.

Even if the tests worked, they'd be useless.

• Problem 2: most changes are statistical white noise.

Everybody acts — the state is shameless about this — as though a change in a district's test results from year to year actually means something. But most of the time it doesn't. Why? Because you're not giving the same test to the same kids — you're giving a similar test to a completely different bunch of kids. When you give a different test to different kids of course you get different results. But that doesn't mean the results tell you anything.

That this year's eighth-grade class is worse at math than last year's doesn't mean the school is doing a worse job — it could be that the kid who sits in the third row and chews pencils is, in fact, as dumb as he looks. Most of the changes from year to year don't mean anything, don't tell us anything except that kids are different from each other, and don't require action of any kind.

And even if they did, we'd never know it, because. . .

• Problem 3: the tests don't provide useful information.

Telling us that this year's class does worse on English exams than last year's is not actually helpful. It's only helpful if there's a reason: It could be that the teachers are slacking off, or it could be that a large number of students who don't speak English as a first language just moved into the district. It could be that a great English teacher is on maternity leave. It could be that an effective program got its funding cut.

Why does this make a difference? Because each situation calls for a different remedy. Saying "Your test scores are down, the state needs to intervene" is like saying "You're sneezing a lot so I'm giving you chemo."

The prescription makes no sense unless there's a diagnosis.

What's the disease? What's behind the test scores? The state doesn't even try to find out. It just starts assigning blame.

There are other serious scientific and technical problems with the tests – but after all the dust has settled the end result is that we have absurd tests providing absurd results that, for some reason, everybody takes seriously.

Here's a better idea: If you want to know how your kid is doing in school, don't wait for the tests that come out next year, ask his teacher today.

If you want to know how your school is doing, look at the college acceptance rates and how well graduates succeed in the job market. Look at whether they participate well as citizens. Find out whether they have some idea of what the world is like.

In the end, isn't that what we care about? You can't test that stuff.