NAEP Test-Taking Pool Grows More Inclusive
Following a push to make "the nation's report card" better reflect the academic performance of all children in America's schools, most states boosted the numbers of students with disabilities and English-language learners who participated in the 2011 reading and math tests that are part of the National Assessment of Educational Progress.
But many states still have far to go to reach the inclusion targets set for them last year by federal policymakers.
Overall, the numbers of 4th and 8th grade students who took NAEP and were identified as having disabilities or being English-language learners rose in 2011, continuing a longer-term trend that began more than a decade ago when NAEP first allowed students to use accommodations, such as additional time, when taking the exams.
But to further drive up inclusion rates for students with disabilities and English-learners, especially in states and school districts that continue to exclude large numbers of such students, the National Assessment Governing Board—the independent body that makes policy for NAEP—set inclusion targets last year for states to meet in the 2011 reading and math exams.
Forty states, plus the District of Columbia, met the goal of including 95 percent of all students in the original testing sample for the reading assessment for grades 4 and 8. On the math assessment, Oklahoma was the only state to fall short of the 95 percent inclusion goal for both grades, while Maryland did so in grade 8.
The rates at which students with disabilities and English-language learners are being excluded from National Assessment of Education Progress in reading and mathematics have declined overall since at least 2003. The goal is for the tests to be administered to 95 percent of all students and 85 percent of special education students or English-language learners who are in the initial sample of students targeted for testing.
"We think states and school districts are taking this seriously and we think there will continue to be improvement," said David P. Driscoll, NAGB's chairman. "We want to see an end to these exclusion rates bouncing all over the place."
Far fewer states measured up when it came to meeting NAGB's goal of including 85 percent of the students identified as having disabilities or being English-learners targeted for testing. And some continue to exclude large numbers of students from the exams.
To ensure that "the nation's report card" is a nationally representative sample of students, the federal testing program selects potential test-takers from a state's entire population at each grade level. State and district educators then may exclude students whose language difficulties or disabilities make test-taking impractical. A state's exclusion rate is the percentage of students from these categorical groups that are removed from testing.
Jumps in Exclusions
In Kentucky, which has one of the highest exclusion rates, 63 percent of the 4th graders who were identified as English-learners in the state's testing sample were excluded from the reading assessment in 2011, up 20 percentage points from two years ago. Among the targeted test-takers in Oklahoma, 60 percent of the 8th graders identified as having disabilities were excluded from the math exam, as were 51 percent of 4th graders. The Oklahoma numbers represent double-digit increases over the exclusion rates in 2009.
Keeping those percentages consistent across states is important because scores could rise and fall with changes in the population of test-takers with learning challenges. The stakes will get higher for states in 2013 when additional rules kick in to further limit who can be excluded from the tests.
Only students with significant cognitive disabilities who take alternate state assessments may be excluded, said Grady Wilburn, an associate research scientist at the National Center for Education Statistics, which oversees the design and administration of NAEP. And for English-learners, school districts will have to include all such students who have been in a U.S. school more than one year. That rule is technically in effect already, but school districts have found ways around it, he said.
NAGB board member Andrew Porter, the dean of the University of Pennsylvania's graduate school of education, zeroed in on exclusions when he joined the board six years ago.
"What we want is fair and valid comparisons over time," he said. "It was obviously an area we needed to look at. I wouldn't have raised it as an issue if I didn't think it was clouding some of the results.
"There's also just the credibility factor: If you have a lot of exclusions, it raises questions," he said. "We want NAEP to be the gold standard."
Even before NAGB approved its policy last year to minimize exclusions from NAEP, the objective had long been to include more students with disabilities and English-learners. That had mostly been done by allowing for a range of testing accommodations for students who needed them. For example, an English learner can take the math NAEP using a bilingual test booklet.
That's why even as the number of English-learners has grown markedly, the percentage of such students participating in NAEP has also increased, said Arnold Goldstein, the director for design, analysis, and reporting at the NCES.
"Allowing for those accommodations has really been a major vehicle to getting those students participating in the test," Mr. Goldstein said.
There are still a few accommodations that NAEP doesn't allow, Mr. Goldstein said. Students can't have someone read aloud to them during the NAEP in reading, for one. Another is giving a test over multiple days, both of which some states allow on their assessments.
But in some states the NAEP accommodations don't appear yet to have had a major effect on inclusion rates for English-learners. Kentucky is one.
The state has roughly 15,500 English-learners in its public schools, out of a total enrollment of 645,000, said Lisa Y. Gross, a spokeswoman for the Kentucky education department.
Ms. Gross said the department had not yet analyzed the 2011 NAEP exclusion rates for English-learners to understand why the percentage had jumped from 43 percent in 4th grade reading to 63 percent. In contrast, the rate dropped in 8th grade reading from 68 percent in 2009 to 41 percent this year. Ms. Gross said one possible explanation for the 4th grade spike is growth in the English-learner population since newcomers don't take NAEP.
Kentucky also allows readers for all parts of its state reading exams. Any student who has received that accommodation on the state test may have automatically been excluded from taking NAEP, Ms. Gross said. But that accommodation window will shrink when the state's new assessment and accountability system takes effect in the spring, she said.
Opportunity to Improve
Laura Kaloi, the public-policy director for the New York-based National Center for Learning Disabilities, wondered if the lack of dramatic progress in national NAEP scores in 2011—performance improved in math and 8th grade reading but was flat at the 4th grade level—is in part because more special education students were included in NAEP this year.
"It shines a light on why students with disabilities weren't being included in NAEP," she said. In a nutshell, schools and districts fear, as they do with other tests, that special education students will bring down scores. Instead, she said, educators should view expanding their testing populations as an opportunity to improve instruction.
The District of Columbia, which had a detailed plan for including more students with disabilities in NAEP, excluded 17 percent of 4th graders and 12 percent of 8th graders in reading, compared with a 68 percent exclusion rate in 2009. But reading scores were almost unchanged compared with 2009 and better than prior years.
"By removing barriers, D.C. was bringing in some capable students who could show their skills and abilities," said Mr. Goldstein of the NCES.
Vol. 31, Issue 12, Pages 1, 14-15Published in Print: November 16, 2011, as NAEP Test-Taking Pool Grows More Inclusive