To the Editor:
“GAO Revises Estimates of Students Excluded From NAEP,” (Nov. 9, 2005) reports that both the U.S. Government Accountability Office and the U.S. Department of Education missed an important error in a July GAO report. But neither the corrected report nor your article deals with the really important question: Are states manipulating their testing programs to look good under the No Child Left Behind Act?
The original GAO report discussed the proportions of students with disabilities that took statewide reading assessments. The report also gave the overall exclusion rate for such students on the National Assessment of Educational Progress.
Because state tests are used to calculate performance under the No Child Left Behind law, states could be tempted to inflate results, either by excluding many students with disabilities, or by manipulating the testing accommodations offered to them. A notable difference between NAEP and state exclusion rates would raise doubts about state testing.
The original GAO report showed that the vast majority of states included large numbers of students with disabilities—over 90 percent—in their “regular” assessments. The GAO also originally reported that the exclusion rate for such students on NAEP was just 5 percent, which seemed compatible with the state data.
But the GAO was wrong. The real exclusion rate for students with disabilities taking NAEP wasn’t 5 percent—it was closer to an alarming 40 percent.
Why do many states include most of their students with disabilities on their own tests, while NAEP does not? Are many states using unreasonable, score-inflating accommodations?
Consider Kentucky, which reads its own “reading” assessment to the vast majority of its students with learning disabilities. NAEP doesn’t allow that.
And given the serious disparity between the corrected NAEP exclusion rate and those of the state tests, will the GAO now probe further?
Richard G. Innes
Bluegrass Institute for
Public Policy Solutions
Bowling Green, Ky.