TESTING THE TESTERS

"It seemed too tricky, even for an S.A.T. question."

So said Colin Rizzio, the New Hampshire teenager who gained national notoriety last week for correctly observing that one of the October SAT math questions was flawed. The uncomfortable nudge Rizzio gave the Educational Testing Service serves as more than a minor corporate embarrassment.

It also testifies to the limitations and fallibilities of any system that purports to be an absolute test of ability. Colin Rizzio scored (before the mistake was discovered) a 750 on the math exam. Indubitably, a small number of students in the country did even better than that. Does the fact that they were better able to answer the formulated questions prove they are more intelligent? Or does the fact that Rizzio had the intellectual flexibility to understand not only how to find the answer to a question, but to intuit how it was designed to work, suggest something?

The mistake Rizzio discovered only came to light because he found a friend who had access to e-mail and sent a query to the Educational Testing Service. Nothing on the test could have measured for the resourcefulness and initiative it took to do that.

Measures of intelligence are inevitably imperfect and contingent on concepts of what "intelligence" consists in. Last week's mishap serves as another reminder of the flexibility of the human mind in evading narrow constructions which would define its capacities. It also challenges those who would evaluate intellect to question, and perhaps to broaden, their criteria.

This week has already brought Colin Rizzio an acceptance from one college to which he had applied. Perhaps, while he is still in the mood for filling out applications, he should look into a position at ETS as well; as his discovery has demonstrated, even those who would evaluate intellect still have something to learn.