March 11, 2001

STANDARDIZING STUDENTS

Peter Sacks was in Denver last week to preach against what he likes to call "the accountability crusade."

The author of Standardized Minds, a tirade against standardized tests, appeared at the Colorado Book Festival.

It was during the testing window for the Colorado Student Assessment Program, and the audience was apparently suffering from a severe case of CSAP fatigue.

State board of education member Gully Stanford was one of the two respondents to Sacks. He made, I thought, a persuasive case that Sacks' arguments might carry weight against the standard sort of test -- all multiple choice, normed on a national sample -- but scarcely applied to Colorado's tests.

Stanford wasn't booed, but the audience clearly signaled by its discontented wrigglings and murmurs that it wasn't at all persuaded.

Just as I was unpersuaded by Sacks.

He believes America is fixated on mental testing because "it is a highly effective means of social control, predominantly serving the interests of the nation's elites."

If that sounds to you like a plausible description of the world, you'll love the book. If not, you'll hate it.

Sacks does have a tendency to construct his demons out of straw. In his talk, he said that Americans are under the illusion that tests have an infallible precision that allows people to be placed in rank order based on differences as small as one point.

But no educator I have ever met believes any such thing. When the Colorado Commission on Higher Education sets the index score (a combination of high school grades and test scores) for the Colorado School of Mines at 110, it isn't saying that everyone who scores 110 or higher will succeed there, and everyone 109 or lower will fail.

It is recognizing that if you have 64 shades of gray, and need to divide them into dark and light, you either must draw a somewhat arbitrary line or maintain that there is no difference between black and white. That's why the index-score rules allow institutions to make 20 percent of their admission offers to students whose scores are lower than the index score.

Sacks asks darkly, "Why is one's father's occupation a better predictor of SAT scores than virtually any other factor?"

Everybody knows why. Smart people (whether identified by IQ tests or not) are much more likely to land high-status, high-paying jobs than stupid people, though we all know of exceptions in both directions. Since intelligence is highly heritable, smart adults are more likely to have smart children. And since the SAT is a pretty good proxy for an IQ test, smart children do well on it.

Sacks rejects every part of that argument. It's not so much that he thinks they aren't true -- he even concedes that intelligence might be 50 percent heritable. It's more that he thinks they (ital)ought(endital) not to be true.

But refusing to acknowledge things you wish weren't true can be folly.

Sacks told the story of a mother from Boise, Idaho, whose son tested at the 99th percentile on the Iowa Test of Basic Skills.

Teachers were in awe of him, his mother said. "Because he did so well on the test, in a way they didn't see (ital)him(endital). They saw him as his test scores."

I thought that was odd. One in a hundred children -- by definition -- is at the 99th percentile, so an average-sized elementary school could expect to have one in every grade, more or less, and several others with scores very close to that. Moreover, my son was someone who tested high but performed indifferently, and his teachers weren't in awe of him. They were furious.

But there's more to the story, as Sacks tells it in the book. The mother is an adherent of the same doctrinaire egalitarianism he is promoting. When she moved to Boise, she said, neighbors told her, "You need to know about this grade school; it has lots of low-income kids, and its test scores are low."

The parents sent their two children to the school anyway. The daughter, who had middling test scores, worked hard and earned straight A's. The son loafed.

"My children represent two very different models of success, but at this point I can't help but feel it's my son who will have a better shot at life success."

Yes, he will, if he too learns to work hard (and most smart people do, eventually). But he'd have learned it a whole lot sooner if his mother had had the sense to recognize that the tests were telling her something important about him.

(743 words)