This letter marshalls a lot of statistics to rebut Standardisto claims.]
Published in Education Week
While I don't want to argue against "Aristotle and common sense," if Herbert J. Walberg wants to make the case that high-stakes testing helps students at risk, then his use of Virginia as an example doesn't help his case ("Accountability Helps Students at Risk," Commentary, April 30, 2003). As other states attempt to emulate the "success" in Virginia, I would, as a parent, advise them to look before leaping
The pass rates on Virginia's Standards of Learning tests have risen over the past five years. In the one case where they have not (history), state policymakers simply moved the cut scores. The practice of narrowing the curriculum and pushing out all things not measured by multiple- choice also improves the pass rate.
While pass rates have been rising, however, other measures of student achievement haven't been quite so definitive, particularly for various racial groups.
The state's average SAT scores have risen for students as a group, but SAT-participation rates have declined. Black males have lower scores than when the SOL tests were first administered.
On the National Assessment of Educational Progress tests for 4th grade and 8th grade math in 2000, the percentage of African-American, Hispanic, and Asian/Pacific-Islander students who performed at or above the proficient level did not differ significantly from what it was in 1990.
The percentage of Virginia public school students scoring 3 or better on Advanced Placement tests has declined from 65 percent in 1997, to 62 percent in 2002. Diploma rates (standard and advanced diplomas) have fallen off as well.
It should be noted, too, that the rise in "achievement" registered by our students on the SOLs has not been matched by their performance on the other test that all Virginia students take, the norm-referenced Stanford Achievement Test-9th Edition. This is especially true in 6th and 9th grades. Between 1999 and 2001, average percentiles in 6th grade reading on Stanford-9 tests did not improve. In 2002, they went up by 1 percentile. In 9th grade reading and math, scores have simply not budged. Compare that with the rising pass rates on Virginia's Standards of Learning tests.
For those who hold up Virginia as an example of rising student achievement, be warned that our definition of achievement is dubious at best. If the goal of education is to narrow the curriculum so that students can bubble-in a few more multiple-choice questions and allow policymakers to pretend students have learned more and better, then Virginia's efforts are paying off.
But almost six years into SOL testing, and with more than $100 million in state money spent on the program, we have rising pass rates on these tests—and not much else.