Pressure from the U.S. Department of Education has led some states to curb a testing exemption that applies only to the 1 percent of students with the most severe disabilities, but districts that have long used that flexibility to win some breathing room in their accountability systems are bristling.
Under the federal No Child Left Behind Act, states are allowed to administer exams based on alternate standards to students with severe cognitive impairments and then count those scores toward their adequate yearly progress, or AYP, ratings—provided the number of scores counted as passing doesn’t exceed 1 percent of the total number of students tested. When more than 1 percent of students take the alternate tests, states or districts must count the scores from exams that exceed the cap as failing. But, for years, the Education Department allowed a handful of states to overstep the 1 percent cap.
A cap was used—rather than a strict definition of who should qualify for alternate tests—so that school districts could use their own judgment in determining who should take the tests, which are not intended for children with moderate disabilities, said Ricki Sabia, the associate director of the policy center for the National Down Syndrome Society, based in New York City. When originally proposed, the cap was one-half of a percent, a proportion she said some research shows is much closer to the actual share of students who might be in need of tests that may hardly resemble current standardized exams.
In guidance published in 2007, the Education Department stopped allowing states overall to exceed the cap, but the department continued to let all states give districts permission to do so and count many more tests as passing without asking the federal government for permission.
Regulations from the U.S. Department of Education allow states to give alternate assessments to some students with disabilities to track achievement under the No Child Left Behind Act. Those assessments, designed for students with significant cognitive disabilities, aren’t based on exactly the same standards that other students in the same grade are expected to meet, but an alternate set of standards generally set by the state. Students who take the alternate tests are selected on a case-by-case basis, although states are supposed to provide guidelines for the teachers and counselors making those decisions.
In some cases, the assessment might be a portfolio of the student’s work from throughout the school year rather than an actual test.
For the purposes of statewide accountability under the No Child Left Behind law, only the scores of up to 1 percent of all students in a district or state—or about 10 percent of students with disabilities—taking alternate assessments based on alternate achievement standards may be counted as passing when calculating adequate yearly progress, or AYP.
School districts may be able to count more than 1 percent of those scores as proficient when calculating their AYP ratings if their states allow it. For example, very small districts, where a few scores could have a large effect on a rating, may be given permission to count more of the scores. Exceptions may also be made if a district has a specialized program for students with severe disabilities that inflates the number of such students beyond what would be typical.
SOURCE: U.S. Department of Education
That states were allowed to continue to give districts leeway on testing bothers advocates such as Ms. Sabia, who worry that kind of flexibility may become part of the Elementary and Secondary Education Act when Congress reworks the ESEA, whose current version is the No Child Left Behind law.
The concern is that giving districts and states carte blanche to test some students with disabilities with exams that are too easy for some of them may deprive students of an opportunity to earn a regular diploma.
There are few hard and fast rules about exactly which students should take the alternate tests. While some students with Down syndrome, autism, and other conditions may need the alternate assessments, others with the same label may not.
“It’s supposed to be a pretty small, small group,” Ms. Sabia said. “We have to be very careful. Not only might [the cap] not go down, there could be a lot of pressure for it to go up.”
In the early years after the No Child Left Behind Act was signed into law in 2002, states struggled with identifying which students should take which exam. At the time, there was no middle ground, said Carlos Martinez, the group leader for standards and assessments at the Education Department. Either a student took an alternate test, which might be a collection of the student’s work and not an actual test, or he or she took the standard exam.
Just four states have developed modified exams approved by the federal government to give to some students with moderate disabilities, and those states are the only ones that can use them.
There are no limits on how many students in a district can take an alternate exam; the individualized educational program, or IEP of a student with disabilities determines eligibility. The cap becomes an issue only if more than 1 percent of students in a given district take the alternate test, requiring the scores of the children who exceed that percentage to be ruled failing, which could damage a district’s AYP rating.
So states began requesting permission to use more scores from alternate tests when calculating AYP, and the federal government relented. Montana, Ohio, South Dakota, and Virginia were granted exceptions; Minnesota’s request was denied.
Ohio’s 1 percent cap was increased to 1.3 percent, an exception that lasted through the 2008-09 school year. In turn, the Ohio education department allowed some districts to raise their own caps to 2.8 percent, a proportion Mr. Martinez called “stunning.”
That percentage translates into roughly 28 percent of all students with disabilities in a given district.
Some districts had asked for 9 percent or 14 percent caps, said Jeanine Molock, Ohio’s director of accountability, though those requests weren’t granted.
‘Time to Evaluate’
“It’s not getting away with something or trying to avoid the accountability consequences,” Ms. Molock said of the exceptions made for some districts. “We wanted to give the field the time to evaluate their students. During that transition, when the feds allowed flexibility, we were going to request it on behalf of our districts.”
She said it was especially important in small districts where 1 percent of students might be a very small number, and even one or two students’ failing test scores could mean a lower AYP rating.
Now, the state is offering even those small districts less flexibility—in some cases, much to their dismay.
The shift, which occurred last school year, is in part because of questions raised by the federal Education Department about Ohio’s practices.
“[The federal department’s] position was always, we needed to notify districts that we need to bring this number down,” Ms. Molock said. Now, Ohio districts with fewer than 1,000 students can count no more than 10 students’ exams as passing when calculating AYP, and larger districts may count up to 1 percent of their enrollment.
In Ohio’s 1,500-student Beachwood city school system, the superintendent is adamant that limiting the number of passing alternate-assessment scores will harm the district’s performance rating in the future.
“Next year, I have been told that we will have 19 students who qualify for the alternate assessment, thus exceeding the cap by nine,” Superintendent Richard A. Markwardt said. “That will have a significant effect on our performance index and probably AYP as well.”
He said his district’s high-quality program for children with significant cognitive disabilities is turning into a liability, and he finds it insulting to those students that some of their scores are automatically considered failing no matter how they do.
“I feel strongly that Ohio’s current practice is an affront to students exhibiting severe cognitive disabilities,” he said. “Furthermore, it damages the reputation of districts that educate large numbers of such students by lowering district report-card scores and thereby jeopardizing local support for the schools.”
Ms. Molock said it’s difficult to tie a lower district rating to just this one part of the formula for calculating those ratings, however, because the formula has so many components, and some of them change from year to year.
The federal crackdown on granting exceptions to states, which in turn has required states to be less generous with districts, was purposeful, Mr. Martinez said. “While exceptions were allowed during the earlier years of NCLB, it’s become rarer and rarer,” he said. “Through training and experience states have learned to better classify students.”
In Virginia, which was once granted a statewide cap of 1.13 percent, the state hasn’t made a request to the federal government to exceed its 1 percent cap in recent years, said Douglas Cox, the state’s assistant superintendent for special education and student services.
Districts that request an exception must meet strict criteria, said Susan Clare, an instructional specialist with the Virginia education department. Those requirements include having 15 or fewer students taking alternate tests in the first place, provided that the number is less than 2 percent of district enrollment and the district has fewer than 1,500 students, among other factors.
Virginia, like Ohio, also makes exceptions for districts with special programs for students with significant cognitive disabilities that might result in an unnaturally high number of those children concentrated in one place.
But Ms. Sabia, of the National Down Syndrome Society Policy Center, takes issue with that approach.
“You shouldn’t have all those kids unnaturally going to school all in one place anyway,” she said.
With so much riding on the alternate tests, including whether students who take them can graduate with a standard diploma, there’s a more basic question at hand, said Kim S. Hymes, the director of policy and advocacy for the Council for Exceptional Children, based in Arlington, Va.
“All of this goes to the direct issue of we need better assessments,” she said.
Efforts Under Way
Those are in the works. The federal Education Department has awarded grants to two groups working on a new alternate assessment that could be used in many states, rather than having each state or district invent its own.
But those groups are separate from the pair of consortia developing tests—aligned with the Common Core State Standards Initiative—for the other 99 percent of students, who include many students with disabilities.
Some of those students, whose disabilities are less severe, currently take a modified version of standardized tests. Those modified tests have raised questions as well, including how to identify which students should take them, and how to deal with the fact that not every state currently has such an exam.
When the new common-core assessments are developed, with a number of adaptations for students with disabilities embedded in them, those modified exams are expected to be unnecessary. The Council for Exceptional Children and other advocacy groups wanted exams for all students with disabilities, no matter how severe, to become developed by the same organizations.
“Have we learned nothing from the current system we’re in?” the CEC’s Ms. Hymes said, referring to the patchwork of tests with varying levels of difficulty, standards, and depth offered across the country now.
But having to even address the difficulties of testing students with significant cognitive disabilities is a huge milestone, said Katherine Beh Neas, the vice president of government relations for Easter Seals, a Chicago-based nonprofit organization that provides a variety of services for people with disabilities.
“One of the good things about all of this is it has forced the assumption that all kids should get the benefit of trying to achieve on content-level standards,” Ms. Neas said.
“These kids typically have been held to extremely low expectations,” she continued. “Before NCLB, we would never even have been able to have the data. It gives us the basis for the conversation, to go back to say, Ohio, and say, ‘Is that really appropriate for all of these kids?’ ”
“If what will get us there is this initial conversation,” she said, “that’s a whole lot better than where we were a decade ago.”
Nirvi Shah, Writer contributed to this article.
A version of this article appeared in the May 11, 2011 edition of Education Week as States Curb Exemptions on Testing