GAO to Ed. Dept.: Improve Peer Review Process and You'll Improve the Tests
The U.S. Department of Education needs to permit direct communication between state education officials and reviewers during the peer review process for evaluations of state assessments and accountability systems, a report by the General Accountability Office concludes. That way questions during the peer review process that arise can be addressed by states quickly. The report suggests that because the identities of reviewers are protected, the Education Department could set up a generic e-mail address for them to use that would allow them to communicate directly with states without their identities becoming known.
The report also says that in cases where the Education Secretary's peer review decisions differ from those of reviewers, the secretary should explain why they differ, so states understand what they need to do to improve their testing systems.
The report brought to mind for me the controversy that resulted a couple of years ago from the peer review process in Virginia. Reviewers found that Virginia school districts were using an English-proficiency test for some English-language learners instead of the state's regular reading test. Federal officials said Virginia had to stop the practice. Administrators of Virginia school districts fought back and then complied with the demand.
Virginia school districts used the regular state reading test for all ELLs for one school year, and after that the federal government approved Virginia's request to use an alternate portfolio test for some ELLs instead of the state's regular reading test.
I've reported on how a number of states had to drop alternate tests for ELLs after peer reviewers said those tests weren't comparable with the regular state tests. And some state education officials were unhappy about having to do what federal officials told them to do because they didn't see the regular state tests as "valid and reliable" for ELLs. The fact that some states were required to stop using alternate tests for ELLs isn't mentioned in the GAO report.
The GAO report touches on some issues concerning alternate tests, such as how they are more expensive to create than tests for regular students, and how they apply to students with disabilities. But the report says little about issues concerning alternate tests for ELLs, except to say that 15 states have created native-language assessments for ELLs.
The report notes that Texas officials say they conduct research regarding how ELLs and students with disabilities can best be included in assessments for the No Child Left Behind Act. But it adds that most states don't have the capacity to conduct such research. The report also notes that because of budget reductions in some states, assessment funds are being significantly cut back.
In sum, the GAO report gives some insight into the challenges facing states regarding the development of state tests under NCLB, but it doesn't provide much new information about the challenges states face to provide "valid and reliable" tests for ELLs.