ROANOKE TIMES

                         Roanoke Times
                 Copyright (c) 1995, Landmark Communications, Inc.

DATE: TUESDAY, September 13, 1994                   TAG: 9409130035
SECTION: EDITORIAL                    PAGE: A7   EDITION: METRO 
SOURCE: ROBLEY S. JONES
DATELINE:                                 LENGTH: Medium


TEST-TAKERS DIFFER

EACH YEAR, the allegation is made that Virginians are not getting enough "bang for the buck" in return for their education dollars because Virginia's SAT scores are below the national average. That seems logical, doesn't it?

Well, not really. In fact, comparing SAT scores of the various states is an "apples to oranges" comparison, for a number of reasons.

First, not all states give the SAT. Many states, especially those west of the Mississippi, give the ACT. The only children in North Dakota who take the SAT are those who plan to attend prestigious East Coast universities that require this test, such as the Ivy League schools. When only a few extremely bright students take the SAT, this gives those states a very high average not reflective of their student populations.

Second, even among SAT states, the percentage of students taking the test varies. Two-thirds of Virginia's students take the test. Less than half take the test in some states. In Virginia, we encourage college attendance and urge marginal students to give the test a try. The more marginal students taking the test, the lower the average.

Comparisons of test scores over the years are tricky as well. The SAT test is not a constant. It has grown more difficult with each revision, as nondiscriminating test items are removed and replaced by tougher questions that better separate the wheat from the tare. The test changes from year to year, so scores may not be neatly compared.

Two tests that all Virginia students take, the Iowa Test of Basic Skills and the Test of Achievement and Proficiency, provide fair comparisons. On these tests, Virginia students did better than their counterparts across the nation on every subtest. Fourth-grade science scores were more than a standard deviation above the mean! Over the past seven years, of the 20 subtests, Virginia scores increased on 13 and declined on only three.

The downside to this good news is that the children of poverty, the children in our inner cities and poor rural areas score significantly below their wealthier counterparts. This problem must be addressed.

Although the SAT tests are given great credence, they are not the best indicator of the performance of Virginia's students. We must all work to improve the quality of education in our commonwealth, but let's base our next step on the most accurate information available, not the average SAT score.

Robley S. Jones is president of the VIrginia Education Association.



 by CNB