July 10, 1991 - Goal Three: Testing

I was 8 years old, shivering at the edge of the Findlay, Ohio public swimming pool. The deep end. In a few moments, I would have to dive in, swim to the marker, float on my back to the count of ten, then swim back.

The first three people before me had been hauled sputtering out of the water by one of the instructors.

Watching me from the side was my grandmother, who after every swimming lesson over the past three weeks, had coached me for another two or three hours. I did not want to disappoint her.

Someone called my name, and I tilted awkwardly into the drink. Swimming Johnny Weissmuller style, head out of the water, I made it to the marker. I flipped over, and just barely stayed afloat for the ten second count. I gulped a few mouthfuls of chlorine, then flipped over again and doggedly inched my way back to the edge.

As I grasped it, I turned to see my grandmother smile, a grin she maintained as the next six kids struggled and sank. Her grandson: the only one who graduated from swimming class.

Some 30 years later, a few things occur to me I didn't think of then. If only one person managed to pass the test, either the test wasn't fair, or the quality of the instruction wasn't very good. One thing was clear: My grandmother taught me to swim, not the swimming instructors.

This story illustrates just a couple of the issues bobbing about in the controversial waters of educational testing.

In some ways, most parents consider the sort of tests their children take in schools much as my grandmother looked at my swimming graduation: tests prove that you have -- or have not -- mastered a subject. But they also say something about the school.

In the face of widespread national alarm about the state of our public educational system, a lot of people have begun talking about the need for standardized national examinations.

That's one of the ideas behind Goal Three of Governor Romer's "Colorado 2000" initiative, which reads: "By the year 2000, all students will leave grades four, eight, and twelve having demonstrated competency in challenging subject matter including English, mathematics, science, history, and geography; and every school in America will ensure that all students learn to use their minds well, so they may be prepared for responsible citizenship, further learning, and productive employment in our modern economy."

But the more I read about testing, the more confusing it all becomes. According to some estimates, schools give about 127 million tests each year. Yet within the educational community, hardly anybody is willing to assert that the tests mean anything definite.

There are two basic kinds of tests: "norm-referenced" and "criterion-referenced." A norm-referenced test compares your performance to the performance of some other group. To go back to my swimming graduation, if only three out of the ten students from the previous year made it all the way to the first marker before they sank, and this year you made it to the first marker and sank, then in a norm-referenced approach, you might be considered "above average." You still can't swim, but you did better than a lot of other people.

A "criterion-referenced" test is blunter: you can do it, or you can't, you know it, or you don't.

Most of the "standardized achievement tests" now used in America are norm-referenced. But few parents understand -- and few schools have explained -- what these tests mean.

The results of such tests utterly depend on the "norm" -- the characteristics of the group to which the students are being compared. The norm might be urban schools (the lowest scores), or rural (middle), or suburban (highest scores).

Or the "norm" may refer to student performance 10 years ago -- which can produce the peculiar result often called the "Lake Wobegon Effect" (where "all the children are above average").

Based on what I've read lately, a national consensus appears to be building toward criterion-based testing. As numerous studies have shown, it takes about 20 hours of instruction to learn how to read; about 100 to learn the basics of reading, 'riting, and 'rithmetic. The rest is practice.

For the basics at least, it doesn't make sense to test on the "bell curve." Either the instruction happened or it didn't. Either the child knows it, or not. It's sink or swim.

Incidentally, the role of the library, in my view, is to provide an opportunity for that "practice" to take place.

Once children get the rudiments of a basic skill -- like reading -- they need to find somewhere to use it in a way that brings them personal, unmonitored satisfaction. There's a difference between practice and drill, which just might translate to the difference between an enthusiastic, lifelong learner and a bored and resentful student.