NAEP To Cut Exams, Test More Disabled Pupils
In a policy reversal, federal officials have announced that next year's National Assessment of Educational Progress will test more students who have disabilities or a limited ability to speak English.
But while the program will be broader in one sense next year, it will not include a previously planned national reading assessment and a state-level assessment in 4th-grade science because of budget constraints.
At a meeting last month, the National Assessment Governing Board, the independent panel that oversees naep, turned down an impassioned request from Sharon P. Robinson, the department's assistant secretary for educational research and improvement, to change rules that have excluded many special-needs students from the Congressionally mandated tests. Board members expressed support for the idea, but argued that its costs were unknown and that funds were in short supply.
In May, the governing board had decided not to offer special accommodations on a routine basis for special-needs students taking the tests. (See Education Week, May 24, 1995.)
But 10 days of round-the-clock work last month by the board, its staff, and Education Department officials yielded a compromise that was officially approved in telephone conference calls on Aug. 15.
The NAEP budget is expected to be tight, and officials had already agreed to forgo other assessment activities: a 1996 study of student transcripts, a larger sampling of private schools, and some math questions on the state-level test that involve extra equipment.
An appropriations bill recently approved by the House would provide NAEP with the same amount in fiscal 1996 as it received in 1995--$32.757 million. That is about $5 million less than was requested by the Clinton administration. The Senate has yet to begin work on a counterpart bill. (See related story, previous page.).
Known as the "nation's report card," NAEP has tested student achievement since 1969. The national-level tests assess a sampling of students in the 4th, 8th, and 12th grades. The first exams designed to yield comparative data at the state level--in mathematics--were launched in 1990.
The governing board and the National Center for Education Statistics wanted to ensure that results from the 1996 NAEP in math and science were comparable with those from past years and to protect the technical integrity of the tests. To change the population of students taking the tests or the conditions under which the tests are given could skew the results and comparisons.
The governing board viewed retaining the 4th-grade math assessment as an especially high priority. Without it, states would lose trend data started in 1992, and the nation would lose the chance to compare its data with scores from about 40 other countries participating in the Third International Mathematics and Science Study.
But Ms. Robinson and Secretary of Education Richard W. Riley were adamant about moving toward a more inclusive sample, a policy that dovetails with the administration's emphasis on setting high standards for all students.
Both the Goals 2000: Educate America ACT and the 1994 law that reauthorized Title I call for greater inclusion of special-needs students in testing, and the administration is pressing for similar provisions as Congress considers the reauthorization of the Individuals with Disabilities Education Act. (See Education Week, May 17, 1995.)
Dividing the Pool
The compromise calls for dividing the national sample of students taking the NAEP tests in math and science, at each of the three grade levels tested, into three subgroups. Each test is expected to involve about 10,000 to 12,000 students, with the sample split equally three ways.
One group will be selected and tested in the same way NAEP had done it in the past.
The second group will be selected using more inclusive guidelines for which students should participate. Those rules had not been written as of late last month, but officials said their intent is to move from rules that allow schools to exclude special-needs students based on subjective judgments to consistent criteria. A decision could come as early as this week, said Gary Phillips, the associate commissioner of the education-assessment division at the statistics center.
A third group of students will be selected based on the more inclusive rules and will also be given any special accommodations they need, such as extra time or one-on-one administration of the tests. Spanish-speaking students whose English is limited will receive a bilingual version of the NAEP tests.
In each state-level NAEP test, students will be divided into two sample groups, selected under the old rules and the new, more inclusive rules. But they will not receive special accommodations, and states are not expected to incur any additional costs.
As part of the inclusion initiative, the Education Department may also spend about $250,000 in 1996 for research on less rigorous versions of NAEP designed for disabled students and those with limited English proficiency. Officials expect to launch a field test of those "targeted" assessments in 1997 and have them fully in place in 1998, Mr. Phillips said.
The cost of the research and of including more special-needs students in the 1996 NAEP could total about $2.5 million, officials said.
Mr. Phillips said the Education Department had agreed to provide financing for those activities, using 1995 and 1996 funds from the budgets of the office of special education and rehabilitative services, the office of bilingual education and minority-languages affairs, and the oeri.
Districts Sign Up
Also last month, the assessment governing board announced that several school districts have applied to participate in the first round of assessments that would provide NAEP results for individual schools.
The Philadelphia, Atlanta, and Milwaukee school districts intend to pay to augment NAEP state samples for math and science assessments in grades 4 and 8. Fairfax County, Va., wants to do the same for the math test for those grades. New York City school officials have said they intend to participate but have not made a formal application.
The decision to drop the state-level data collection in 4th-grade science could mean that those districts would have to spend more money to supplement their samples from the national test.