|The critics of the Bush plan want to preserve the status quo, and they have been effective.|
George W. Bush began his presidency with a good plan to improve education. He proposed that the states should test every child from grades 3 to 8 in reading and math, to make sure that parents and teachers know how well each child is doing. The federal government, he said, would pay for test development and would recognize those states that raised student achievement. With this strategy, he argued, no child would be left behind.
Integral to the Bush plan is the idea that each state would select or develop a test that produced comparable grade-by-grade data from year to year, and that the states’ progress (or lack of progress) would be independently confirmed by the federally funded National Assessment of Educational Progress, which has been testing samples of students in the states and the nation for more than 30 years. President Bush modeled his plan on the success of the Texas testing-and-accountability strategy, which has enjoyed strong bipartisan support over the past dozen or so years.
On NAEP tests of mathematics, Texas 4th graders went from 21st in the nation to sixth in the nation from 1992 to 1996. No less impressive, black 4th graders in Texas outperformed their African- American peers in every other state in math. On the NAEP writing test for 8th graders in 1998, Texas was one of the top-scoring states. Twenty percent of Texas’ black and Hispanic students scored “proficient” on the writing exam; for Hispanic students, that was twice the national average, and for black students it was nearly three times the national average for their group.
Opponents demand that states be allowed to use inconsistent tests and to “confirm” their states’ progress with less reliable alternatives to the well- established NAEP.
Despite this impressive demonstration of the success of Texas’ accountability program and of the value of using NAEP as independent confirmation of the state’s claims, critics have been dismantling the key elements of the Bush proposal. A coalition of opponents from across the political spectrum demands that states be allowed to use inconsistent tests and to “confirm” their states’ progress with less reliable alternatives to the well-established National Assessment of Educational Progress.
Some conservatives in Congress insist that states should be allowed to use a variety of tests, not just a coherent state test that produces comparable results. They have also maintained that the states’ claims should be confirmed by a test of their choice, rather than NAEP. Commercial test publishers, protecting their large niche in the marketplace, support the conservatives’ demands to preserve local control. The National Governors’ Association and the Council of Chief State School Officers, pledging allegiance to flexibility, also reject the prospect of using a state-designed test that would produce comparable data annually on the performance of each student; they recommend a system of state and local tests, not necessarily comparable, so long as all children take some test.
The critics of the Bush plan want to preserve the status quo, and they have been effective. The bipartisan bill under consideration in the Senate would allow states to use a variety of tests, not exams that yield “comparable results.” A House Republican bill undercuts both parts of the Bush plan, first by not requiring comparable state tests and second by allowing states to confirm their state tests results with an exam other than NAEP.
|Many of those who recommend alternatives to NAEP are fearful of a national examination and worry that the U.S. Department of Education will create a national curriculum.|
Why should states use a test that is comparable from year to year, rather than a variety of noncomparable tests? The power of the Bush plan, which even the administration has not made forcefully enough, is that schools can use regular assessments of each student’s academic progress for diagnostic purposes. By comparing the results of each year’s tests to previous years’ results for individual students, school leaders can determine which students are learning and which are not. They can see the “value added” each year for each student and can use this information to identify unusually effective teachers, as well as teachers in need of help. Parents can see clearly how their children are doing in school. Proponents of multiple tests claim that noncomparable exams can be “linked” or “equated,” but testing experts say that this is not feasible.
Why should NAEP be used to confirm the states’ claims about academic gains? NAEP’s credibility is unparalleled. In its regular reports on academic progress, NAEP has proven to be reliable and authoritative. What is more, NAEP is the only testing organization that can provide an independent audit for the states. No current standardized test can serve this purpose. Commercial test publishers sell the booklets for their standardized exams to states and districts, where they are reused for five or six years until new editions are published. The administration and security of these testing programs are controlled by states and districts, not by an independent external organization.
Many of those who recommend alternatives to NAEP are fearful of a national examination and worry that the U.S. Department of Education will create a national curriculum. But the great strength of NAEP is that its content, standards, and policies are controlled by an independent, bipartisan governing body, the National Assessment Governing Board.
Congress created this board in 1988, when it authorized state-level administration of NAEP. The Education Department, however, still manages many of NAEP’s operations and reports. Now is the time for Congress to clear up all ambiguities by giving the independent board full control of NAEP operations.
The power of the Bush plan is that schools can use regular assessments of each student’s academic progress for diagnostic purposes.
The Bush plan does not threaten local control. It proposes to allow each state to select and control its own assessment, then to get a second opinion from a highly respected source: the National Assessment of Educational Progress. Claims about “local control” are a red herring. Presently, districts across the nation are not writing their own tests. They are taking state tests and nationally standardized tests, none of which was written in the local school district.
If the current insistence on noncomparable tests is enacted, the great promise of the Bush education plan will be lost. Then we will have testing for the sake of testing, which will rob the initiative of its purpose. At great expense, we will have mountains of data, of no comparative value. The status quo will be reaffirmed. And count on it: Lots of children will still be left behind.
Diane Ravitch is a nonresident senior fellow at the Brookings Institution in Washington. She was the assistant U.S. secretary of education for research in the first Bush administration and is a member of the National Assessment Governing Board.
A version of this article appeared in the May 02, 2001 edition of Education Week as The Travails of the Bush Plan For Education