Every year, the American public spends hundreds of millions of dollars on testing our students. Yet we find out very little about what they know and are able to do. What do I mean? After all, students take their standardized reading and math tests, school districts announce the results, and we hear that "This year 56.7 percent of our students are above average in reading and 62.2 percent above average in math, whereas last year..." Sure, but what, exactly, are those numbers telling us?
Dr. John Jacob Cannell found that most states report the majority of their students to be "above average" on standardized tests. Impossible? No, because in the newspeak of standardized test reporting, "above average" doesn't mean in comparison with the kids who took the test that year. It means above average of a sample group of students who took the exam five, ten or more years ago.
Still, shouldn't we still applaud when scores go up? Probably not. The higher scores are not necessarily a sign that schools are improving, that students know more reading and math. They usually just mean that more and more time has been spent on practicing answering the kinds of questions that will be on the tests. Kids may actually be learning less because other, more important things that won't be tested are being neglected.
The next time the test results are publicized, instead of being dazzled -- or disgusted -- by the numbers, stop and ask yourself what the scores mean. If 58.6 percent of ninth graders are above average, what can they do? Can they read a newspaper? Can they write a good letter? Can they extract information from an almanac, graph or chart? This is the kind of information that's understandable and useful to the public and policymakers. It's what we need in order to know how well or how poorly our schools are doing.
The National Assessment of Educational Progress (NAEP) regularly supplies this kind of information from a national sample of students. But NAEP was prohibited from reporting results for individual states, school districts, schools or students because, when it was put into place by the Congress in 1969, it was feared that this move would require testing so many students that it would lead to a national curriculum and federal control of schools.
In spite of these limitations, and perhaps partly because of them, NAEP has been able to give us useful, and often sobering, information about what American students know. It has told us, for instance, that only 2.6 percent of our 17-year-old high school students - the dropouts aren't counted -- can write a good persuasive letter, only 20 percent can write a simple two-paragraph letter applying for a job, only 12 percent can arrange a series of simple fractions in size order and only 3 2 percent know in which half century the Civil War occurred.
But national averages just don't move people to action because they don't tell them how their own school systems or kids are doing. So last year, a national study group headed by former Tennessee governor Lamar Alexander and H. Thomas James, president emeritus of the Spencer Foundation, recommended that the prohibition on how NAEP results are reported be loosened a bit. NAEP has now expanded so that results for participating states can be broken out state by state though not by school district, school or student.
State officials, taxpayers and parents will now know how the performance of students in their state compares with students in other states, but that still won't tell them what they need to know. Whatever the state results, districts and schools will be able to say, "That's not us. We're much better. That district down the road or at the other end of the state brought down the score." Of course, giving states a chance to compare their current and future results will provide some useful information over time. When taxes are cut, do students achieve less? If a state adopts a particular set of texts or standards for becoming a teacher are there changes in student achievement?
But that still leaves an unfilled need for assessments for people in local school districts. A further expansion of NAEP to meet that need may be the answer. Or it may not be. NAEP is still mainly a multiple choice test. And it may be unwise to expect NAEP to do something it wasn't designed for.
Does that mean we have to continue to rely on those idiotic standardized tests? Fortunately not. New tests now in the works assess student performance by asking students to do such things as carry out real science experiments, write essays based on their own research and make oral presentations. Instead of being boxed in by the one-right-answer format of a multiple-choice test, kids can explain their reasons for answering questions in a particular way. They can show what they know and are able to do.
Imagine the difference if tests like these drove the curriculum instead of low-level, multiple choice tests. Teachers would be preparing kids for tests by helping them learn how to think, write and solve problems skills they need in the real world. Connecticut, New York, Vermont, California, Michigan and a handful of other states are pioneering tests like these right now. Why just a handful? The present idiotic system of assessment has got to go. We need a way of knowing, really knowing, how our students are doing.