Los Angeles Times, July 12, 1996, Friday, SECTION: Metro; Part B; Page 9; Op Ed Desk

Measuring Political Correctness;
Knowledge: Sneaking in an Evolution Question
Makes the Literacy Test Really a Litmus Test.


by Mark Hartwig

Do you think that human beings, as we know them today, developed from earlier species of animals? No? Then that's a sign that you're scientifically illiterate--at least according to the National Science Foundation.

In May, the NSF released the results of a brief survey designed to measure what Americans know about basic science. The survey, which purported to test knowledge of simple scientific facts, included questions about whether the center of the Earth is hot, whether the oxygen we breathe comes from plants and whether light travels faster than sound.

Mixed in with those questions. however, was one asking if humans had descended from earlier species of animals. As with the other questions, the "correct" answer was treated as if it were obvious and noncontroversial--a matter of mere factual knowledge.

That item, however, measures belief, not knowledge. Most Americans understand that humans are thought to be descended from other species. But a great many of them would answer "false" because they just don't believe it. You can call these people dissenters, you can even call them heretics, but you can't say they lack knowledge unless you redefine the word "knowledge." In testing jargon, the item lacks validity. That is, it doesn't measure what it purports to and can only mislead if used as the basis for any kind of policy decisions.

But maybe that's what the NSF wants. It's unlikely that the question is simply an innocent error. Professional test makers employ elaborate procedures to ensure that every item measures just what it purports to--nothing more, nothing less. Although test makers seldom achieve such purity, it defies belief that such a blatant error could have made it through any reasonable validity check. Sneaking a cannon through airport security would be easier.

Most likely, the item was designed to label unbelievers as scientifically illiterate. Instead of giving people a fair chance to demonstrate their knowledge, it forced them to choose between an obvious "fact' and their own beliefs--between "literacy" and conviction. What the NSF wanted was not a literacy test but a litmus test.

Mark Hartwig is managing editor for the Foundation for Thought and Ethics, based in Richardson, Texas. He worked for several years as an evaluation specialist at UC Santa Barbara and has served on an NSF science testing committee