The board that oversees the National Assessment of Educational Progress has revised its rules to encourage local districts to participate in next year’s testing program.
The National Assessment Governing Board has extended the window in which participating schools can administer the NAEP science and math exams in 2000.
At its Nov. 20 meeting in Washington, the governing board—known as NAGB adopted a policy allowing states to offer the tests the last week of January and the first week of March as a way to entice them to take part. Before the action, the board required states to give the tests during the first four full weeks in February.
The National Center for Education Statistics, an arm of the U.S. Department of Education, has campaigned to include all 50 states in the 2000 NAEP samples and had recruited 48 states to participate—the highest number since the federal testing program began offering state-by-state results in 1990. Only Alaska and South Dakota declined to join. (“Record Number of States Commit to NAEP,” Oct. 6, 1999.) While many states are willing to give the exams, school districts are hesitating because the federal testing program coincides with the administration of many state tests and takes away instructional time.
Because NAEP is a sampling of student results, it doesn’t require every student in a school to take the test. But it does require a representative sample of students, which can disrupt several classrooms by pulling boys and girls from each of them.
If a state can’t recruit enough districts to provide a 2,500-student representative sample, the NCES will not be able to produce an accurate gauge of student performance.
In its resolution, NAGB also granted Gary W. Phillips, the acting NCES commissioner, the power to grant waivers to states beyond the extra two-week window.
The national assessment will collect state-by-state results for 4th graders and 8th graders in science and mathematics next year.
It also will give its 4th grade reading test to a national sample of students, but won’t produce state-by-state results because Congress only appropriated enough for national scores.
While the governing board’s vote was unanimous, NAGB member Thomas H. Fisher, Florida’s director of student-assessment services, abstained because his state may seek a waiver under the new policy.
Districts on Their Own
As states struggle to collect enough participants for next year, at least five districts are potentially eligible to receive their own NAEP scores simply because they supply such a large portion of their states’ samples, according to statisticians who work on the testing program.
To receive NAEP scores similar to a state report, a district must have 20 percent of a state’s 2,500-student sample, according to Richard Valliant, a senior statistician for Westat, the Rockville, Md., research company that designs NAEP’s complicated sampling methods.
In 2000, Albuquerque, N.M.; Chicago; Clark County, Nev.; New York City; and the Christina district, which serves Wilmington and Newark in Delaware, would qualify. Had Alaska chosen to take part, Anchorage would have qualified as well.
But NCES officials won’t be offering scores to those districts. At its August meeting, NAGB suspended its policy of offering district- level reports until Congress passes legislation clarifying that both state and district officials must agree to the public release of the scores. Earlier this year, New York state officials wanted to know New York City’s scores, but city officials questioned their validity.
“We have a policy in flux,” said Michael J. Guerra, the executive director of the National Catholic Educational Association in Washington and the chairman of NAGB’s reporting and dissemination committee. “We said: ‘It’s desirable; let’s do it.’ But it’s not as easy as we thought. We haven’t had the opportunity to think through that second set of questions.”
