SOL Testing

Virginia's Standards of Learning (SOL) program is a little like the hotel bathroom scale that says you're five pounds overweight even though you know you've been watching your diet. Sure enough, when you get home, you find you haven't gained an ounce.

When the results of the first round of SOL tests were made public a little over a year ago, fewer than 3 percent of the state's public schools met the accreditation standards imposed by the state. In Fairfax County, where we are proud of our schools and the performance of our students, only 7 percent of the schools had met the standards.

Like the traveler on the hotel bathroom scale, we immediately ran home to look at our results on such measures as the College Board's SAT program and the nationally standardized Stanford 9 exams. With a sigh of relief, we observed that our school system had not suddenly changed overnight. Like the traveler, we concluded that the problem was not with our scale but rather with the state's.

Virginia is looking forward to the third administration of the SOLs. Last year the statewide percentage of schools "passing" went from 3 percent to 7 percent. In Fairfax County, we went from 7 percent to 20 percent. It is fair to say that we can expect improvement, but in all probability, the percentage of schools passing statewide will not exceed 20 percent. Is this an accurate reflection of the status of education in Virginia? We don't think so.

The tests are reliable. The state has proven that: They show the same awful results every time they're administered. It's just that the tests, like the scale, are not calibrated properly.

I was asked to serve on the panel that oversaw the establishment of the "cut points" for the tests. In nearly every case, the Board of Education adopted cut points that were in the high range of what the committees had recommended. In some cases the cut points exceeded the recommended range. This was a political decision, not one based on the advice of those who had been asked to calibrate the tests. The bar was set unreasonably high.

As a result, educators and parents throughout the state are becoming increasingly concerned about Virginia's high-stakes standards. The freshman class entering our high schools this September will be the first group of students required to pass SOL tests in order to receive a high school diploma. A recent article in The Post mentioned that many parents are now placing their children in private schools, which do not give the SOL test, for fear that their children might otherwise be denied a high school diploma.

Teachers also complain about the pressure they feel to teach to the test and to have their students pass the SOLs. Real estate firms market neighborhoods based on the performance of the local schools on the SOLs. By the year 2007, school accreditation will be based primarily on whether 70 percent of the students have passed the tests in each of the subjects tested. Understandably, there is a growing backlash around the country against this sort of high-stakes testing.

The problem is not with the quest for higher standards and challenging instruction. The stumbling block is the tests. Testing experts will tell you that no single, group-administered standardized test by itself should determine whether a student receives a high school diploma. Yet in Virginia, 12 such tests can do just that. Never mind the final grades on a course that the student has participated in for the past 10 months or performance on other recognized, valid and reliable exams. If the student fails the social studies SOL, forget it--no cap and gown. And if that failure brings the passing rate for the school down to 69.9 percent, no accreditation.

Virginia's Board of Education has been receptive to suggestions for change. Board President Kirk Schroeder takes in all the criticism and does his best to implement the good suggestions. The board has addressed issues dealing with student mobility and with students with limited English proficiency. A proposal is on the table to offer a "basic" diploma for learning-disabled students, who might not otherwise be able to receive any kind of diploma.

But the solution to Virginia's SOL problem will not be obtained piecemeal. The core of the problem remains. The tests, in and of themselves, should not be the sole determiners of scholastic success and school accreditation. They could be used to monitor performance and gauge improvement, of course. And we could use them as part of accountability models such as the one we have recently implemented in Fairfax County. But we should never use such a test as the sole criterion for deciding a child's future or the educational worth of a particular school.

The writer is superintendent of the Fairfax County Public Schools and a member of the National Assessment Governing Board.