Last week, Education Secretary Lauro Cavazos showed off the 1990 wall chart, the latest in the Education Department's yearly compilation of facts and figures about how our schools are doing. The signs Secretary Cavazos pointed to looked bad.

Some of the bad news was that the number of students graduating from high school is down ( from 71. 7 percent in 1987 to 71.1 percent in 1988). So are the scores on exams taken by students applying to college. (The average score on the American College Testing program exam [ACT] declined from 18.8 out of a possible 36 in 1988 to 18.6 in 1989; and the average score on the Scholastic Aptitude Test [SAT] declined from 904 out of a possible 1600 in 1988 to 903 in 1989.)

But Secretary Cavazos didn't tell us what all these facts mean. Is a 0.6 percent drop in the graduation rate significant or is it a statistical blip? Are more kids dropping out because the higher standards are too tough for them or is it because the system isn't challenging enough? The wall chart lines up the facts and figures, but Secretary Cavazos didn't seem to think it was necessary to give us any interpretation of the facts.

There were plenty he could have offered:

♦ For instance, he could have told us that the reforms in place now are the right ones, but it will take more time for them to catch on. If we're patient, we'll see wonderful results in a few years.

♦ He could have said that the reforms are good by they are being sabotaged. (And here he could have joined his predecessor William Bennett by blaming unions for the sabotage or he could have broken new ground by blaming principals, school boards or even kids.)

♦ He could have said it's too early to tell anything about the changes we've been making in the schools. After all, the kids who are graduating this year learned their basic skills in the late 1970s -- a number of years before the current school reform movement got under way.

♦ Or he could have told us that the things we're trying now are dead wrong and will never work. Then he could have offered us some suggestions of his own.

Probably there are other ways of explaining the rather depressing-looking collection of facts on the wall chart. The point is that facts don't speak for themselves. They are only the raw materials on which we can base a conclusion -- or a number of possible conclusions. And we won't ever decide anything useful about our education system if we just look at education statistics the way we look at daily fluctuations in the stock market.

In any case, there is a limit to what could be gleaned from these figures -- even if we were prepared to interpret them to within an inch of their lives - because what the wall chart tells us is abstract, impoverished and misleading. For all the thousands of facts it presents, it can't give any useful information about how our students are doing.

Suppose we know that a certain percentage of students who entered high school in 1985 graduate four years later. What does that statistic tell us about how well those students can do two-step math problems, write letters of application or read materials beyond what they find on cereal box? What does the difference between scores of 18.6 and 18.8 on the ACT translate into? These are measures, but they don't measure what we want to know. They don't five us any concrete sense of our students' skills or limitations.

If we really mean business about reforming our education system, we need better information about what and how much our students are learning. And instead of depending on a lot of statistics that were never designed to be measures of student achievement, we need a system of assessments that will give our students a chance to demonstrate what they have accomplished.

Now is a good time to think about that kind of system. The national goal set by President Bush and the governors at the Education Summit last September included ones for student performance in various academic areas. Assessments that tell us how well students are meeting those goals will give us valuable information about exactly how well our schools are doing, too.

Well-designed tests will cost money to develop and to administer. But we'd be very foolish to skimp here. Our education system nationwide is a $200-billion enterprise, and we know its productivity is very poor. We need a sensitive way of measuring improvements -- or decreases -- in student learning.

A system of assessments like the one I've described could do this; the wall chard will never be able to.