One of the national education goals adopted last year by President Bush and the governors is that U.S. students will be "first in the world in math and science" by the year 2000. Now the latest math test scores from the National Assessment of Educational Progress (NAEP) are in. These scores tell us that only 5 percent of our graduating twelfth graders can solve problems using algebra and geometry, and half of those about to graduate from high school cannot even do what NAEP calls seventh-grade work.

It's not the first time we've heard about the dismal performance of U.S. students in math: NAEP has been giving us the same bad news for 20 years. So it should come as no surprise that our kids are nowhere near "first in the world." What can we expect now? Only, I'm afraid, that another test will be administered two years from now to see if we're doing any better.

Suppose you were feeling terrible and went to a doctor who tested you, said you had a fever and asked you to come back in two weeks. And suppose you returned every couple of weeks for a year or so to be tested, and every time, you got the same bad news but still no diagnosis and no advice. Few of us would think much of a doctor who did nothing but test us and give us bad news. But that's what's been going on in education for 20 years.

Student math scores haven't changed because we haven't been doing anything between the tests to improve them. Testing is necessary but it's not enough, and no president or governor should be praised for simply putting in a system of tests and reporting on the miserable results. The real question is, what can be done to improve the results?

Along with test results, we should get a diagnosis. It might suggest that the problems leading to poor math achievement start in elementary school. Many teachers there have had little or no education in math so they can't do a good job of teaching it to children. However, replacing them is impossible because we don't have any better-qualified people waiting for their jobs. What can we do? We might pair up teachers who are strong in math with teachers who have other strengths and arrange for them to exchange classes. So an excellent math teacher might teach an extra math class in return for having a teacher who was especially strong in language arts take over that class. As a result, children would be more likely to have competent teaching in math. It would also help if we made more creative use of technology and offered programs to help teachers whose knowledge of math is inadequate.

But there are other possible diagnoses. Many people teaching math in secondary school are not math teachers at all: Only 6 out of 10 of them have degrees in math (or math education). What to do about this? Why not use incentives to increase the number of qualified people in the field? For example, offering math teachers higher salaries would encourage prospective teachers to prepare themselves to teach math, and it might even lure some former math teachers back to teaching. Or we could offer bonuses to teachers of other subjects if they became qualified in math.

We could consider reorganizing the job of teaching math to make more effective use of our qualified teachers. Suppose outstanding math teachers headed up teams. They would be in charge of classroom instruction, and other team members would assist them by coaching students, teaching in areas where they were strong, grading papers and so forth. Besides making sure that students got excellent instruction, this system would also help teachers assisting on the team to improve their own math teaching.

We could also consider testing teachers before they were hired. I'm not talking about the kinds of low- level tests many teachers now take -- and pass -- as a condition of employment. I mean tests that would guarantee that prospective math teachers had solid knowledge of what they would be teaching.

The poor results may also be due to the fact that few high school youngsters do much work. They know they'll graduate from high school and, if they want to, get into college -- even if they'd have trouble doing seventh-grade math. How can we encourage them to make a real effort? We could say that kids who don't achieve at a certain level can't get loans to attend college- or even be admitted to a four-year college -- and that generous scholarships will be given -- to high math achievers. We could ask the businesses that employ high school students to give after-school jobs only to kids doing well in math. We could require students who haven't reached a given level in math by seventh or ninth grade to attend summer school to catch up.

Probably all of the things I've mentioned contribute to the terrible performance of our students in math and probably there is no single solution to the problem. But one thing is sure: Our math results are a disaster. Secondary schools in other industrial countries are graduating from 16 percent to 30 percent of students who know more math than our top 5 percent. Find that shocking? I do. But it's just as shocking that we've known this for over 20 years and have been working under the assumption that all we have to do is to give the patient the bad news periodically and be will somehow find a cure for himself.