As published in The New York Times, October 27, 2000
WASHINGTON — Earlier this week the Rand Corporation, a research group, released a study that questions rising student test scores in Texas. According to the study, Texas students demonstrated greater improvement in recent years on the state’s own exam than on the leading national test, the National Assessment of Educational Progress.
These “stark differences,” according to the study, written by Stephen P. Klein and others, raise serious questions about whether the state test is a meaningful measure of academic progress.
Predictably, the Gore campaign claimed that the new study “devastates” Gov. George Bush’s education claims in Texas. But to draw that conclusion from this flawed and misleading study is wrong.
First, the 14-page study is relatively unsophisticated and is based on incomplete data. “Understanding the source and consequences of the impressive score gains” on the state test, Mr. Klein himself writes, “would require an intense independent study. We have not done that.”
Moreover, a more complete study, released last July, contradicts Mr. Klein’s findings. David Grissmer, also a researcher at Rand, released a 250-page report that was far more rigorous and comprehensive than Mr. Klein’s. Among Mr. Grissmer’s findings: Texas students ranked high nationally, particularly in fourth-grade math, and black fourth graders in Texas made bigger gains than fourth graders in any other state on the math portion of the national test.
It is noteworthy that Mr. Grissmer has taken issue with the Klein report. “I continue to support our conclusions” that the national test score increases “were among the highest across states,” he said in a statement.
What about the gap in scores between the state tests and national tests? Mr. Klein neglected to provide some vital context. It is normal for state tests to show better results than national ones. There are straightforward reasons for this: state tests are more narrowly designed; they test more basic skills; they intentionally align themselves to the state standards and curriculums (which national tests do not); and they provide more incentives, like grade promotion, for students to do well.
“It is premature, based on the current evidence, to conclude that there is something amiss” in the state testing program, Mr. Grissmer said.
The increase in state scores may be greater than that in national scores, but the fact remains that according to the most recent national test scores, black and white fourth graders in Texas performed best in the nation, while Hispanic fourth graders placed sixth.
To put it another way, even if state scores are discounted, Texas should still be praised for making gains, according to national test scores. Numerous impartial scholars, analysts and observers have attested to the state’s achievements in education. Even the Education Trust, a liberal advocacy group, has praised the Texas education system for its improvements in math and writing. According to Kati Haycock, the group’s director, the new Rand study “offers an incomplete and misleading picture of the performance of Texas students” on the national tests.
To the degree that this study draws attention to Governor Bush’s record in Texas, we believe it should work to his advantage. He can rightly claim very impressive accomplishments in the area of education; the closer people look at the Texas record, the better it is. Nothing in the latest report contradicts this.
William J. Bennett served as secretary of education under President Reagan. Chester E. Finn Jr. is a senior fellow at the Manhattan Institute.