Big Jump in Test Scores. Probable Fraud.

This isn't about Tucson, or even Arizona. It's about an elementary school in Charleston, South Carolina. But the story, which is initially encouraging, then potentially disheartening, could happen anywhere.

Here's the basic story. New principal takes over an elementary school in an impoverished neighborhood. She cares, works long hours, loves the kids, helps families when their food stamps run out. And from the first year, test scores rocket upward.

That's the encouraging part.

Then someone notices that students' tests have an unusual number of erasures on their answer sheets, and most of the new answers are correct. So the next year, the tests are given in a closely monitored situation, and scores go down. The principal, who said she did nothing wrong, moves to another district and will not comment on the situation.

That's the disheartening part.

The verdict isn't in yet, but I suspect someone went over the students' answer sheets and made enough changes to increase the scores. It's easy. I could have done it with my students' state reading tests, since I collected them after the students finished and turned them in at the end of the day.

For that matter, I could have pre-cheated. I had the test booklets in my hands days before I handed them out. A little judicious "teaching" that covered certain questions I knew were on the test would have given my students' scores a bump.

I didn't pre-cheat or post-cheat, by the way. But I certainly coached my students to raise their test taking abilities, which gave them higher scores than they would have gotten otherwise. Perfectly legit. If I happened to be a better testing coach than the teacher next door, my students would score better even if the skill levels of the two classes were identical.

Don't trust standardized test scores. There are too many ways, legitimate and otherwise, to manipulate them.

There's a big push for teacher merit pay based on students' test scores. Do you honestly think lots of teachers won't do whatever they can to raise scores if it means a $4,000 raise? Of course they will. Teaching directly to the test — and disregarding non-tested curriculum — will increase. Some teachers will change an answer here and there. Principals will get involved as well, because high scores make them look good.

High stakes tests, especially if they have some financial award attached, distort and potentially corrupt education. There's nothing wrong with the judicious use of standardized tests, but we've given them an importance they shouldn't have.

But here's my real disappointment. I would love to believe that a principal who worked her tail off, loved and cared for the students and got the staff on her side could work miracles. I'm sure it happens sometimes, though "progress" is more likely than "miracles." But it doesn't happen often, and when it does, attempts to recreate the progress elsewhere tend to be unsuccessful.

Yeah, I know I sound cynical. And to some extent I am. But wrapped in my cynicism is a ray of hope. We have to realize that schools won't change things on their own. Students raised in poverty without much educational stimulation or parental support in their homes won't turn into scholars because schools discover some magic educational bullet. If social and economic opportunities in a community improve, if parents and the community in general make their children's education a top priority, students will learn more in school and test scores will go up. That's the only way real progress will take place. If we expect schools to do it on their own, we're going to continue to be disappointed.