'Read Aloud' Assistance on Common Tests Proves Contentious
Accommodation for students with print-related disabilities has assessment consortia taking differing approaches
Faced with the decision of whether to allow students with dyslexia and other print disabilities the option of having text passages on the common-core tests read aloud to them, the two federally financed consortia responsible for creating the general assessments took a Solomonic approach.
Rather than prohibit the so-called "read-aloud accommodation" entirely or allow reading aloud with no restriction, the Partnership for Assessment of Readiness for College and Careers decided to permit text passages to be read to students, with a notation on score reports saying no claims can be made regarding the student's foundational reading skills. The Smarter Balanced Assessment Consortium opted against the read-aloud accommodation for students in grades 3-5, saying it would invalidate the language constructs being measured; students taking the test in higher grades may use that accommodation.
Like many decisions that attempt to strike a balance between two opposing philosophies—those who wish to leave the use of read-aloud accommodations to a school-based team and those who want to prohibit its use entirely—the decisions of the testing consortia appear to have left no one entirely happy.
And now critics on either side say the consortia's decisions on reading aloud could be setting districts up for violations of special education law, or could ultimately leave a swath of students unable to read because the read-aloud accommodation was used as a crutch.
Richard Allington, a professor of education at the University of Tennessee and one of the country's most recognized experts on early literacy, calls the accommodation "cheating."
"What special education does best is create illiterates," Mr. Allington said. "I know why they don't want their kids tested on reading activity. It's because they've done a terrible job of providing those kids with high-quality reading instruction."
But special educators believe just as strongly that for some children, a read-aloud accommodation is the tool they need to demonstrate what they know.
Lindsay Jones, the director of public policy and advocacy for the National Center for Learning Disabilities, said she was "stunned" that Smarter Balanced chose to ban read-aloud accommodations for elementary students, even for visually impaired children who may be in the early stages of learning Braille. (An official with the consortium explained that the testing group was told by its legal experts that it could not open up an accommodation to blind children without opening it up to all students with disabilities.)
The learning-disabilities center has argued on behalf of restricting read-aloud accommodations only on test items that gauge print decoding and fluency. Questions that measure other literacy skills, such as picking out a main idea of a text passage, should be open to the accommodation.
"There's been a lot of conjecture and anecdotes about evidence that states overuse read-aloud. I don't know why we would let that dominate our conversation going forward," Ms. Jones said. "Why can't we create items that will test comprehension, that will test decoding? That is a failure, to me, of the promise of these exams."
The restrictions also could represent a conflict with the Individuals with Disabilities Education Act, which gives broad powers to school-based teams of teachers, administrators, parents, and others who have deeper knowledge of a child's capabilities to determine what accommodations may be necessary. Perry A. Zirkel, a professor of education and law at Lehigh University in Bethlehem, Pa., said that the approach taken by Smarter Balanced could be more legally sound than drawing attention to an accommodation. The Educational Testing Service, which administers the SATs, was sued by a student for its policy of flagging tests that were taken under an extended-time accommodation. The testing company agreed to end flagging in 2001.
"My prediction is that [PARCC] thinks it's getting out of the problem," Mr. Zirkel said. "If this gets challenged, they will likely follow the ETS approach."
Since 1990, the National Center for Educational Outcomes, based at the University of Minnesota-Twin Cities, has been tracking accommodations policies for students with disabilities. It found that while the read-aloud accommodation on test reading passages may be controversial, it is not rare. Several states currently allow text passages to be read aloud to students with certain disabilities, though in some cases, that accommodation results in the test being invalidated for accountability purposes.
States currently vary on whether they allow text passages on state tests to be read aloud to students. Some prohibit this accommodation, while others allow it in certain circumstances, such as for students who are blind or visually impaired. Some states allow it with implications for scoring; for example, the test may be invalidated. Other variables are whether states allow human readers, software-based text-to-speech, prerecorded audio, or some combination of these.
Kentucky is an example of a state with an expansive read-aloud policy, "on the premise that the intent of reading is to measure comprehension." Hawaii's guidelines for the read-aloud accommodation are much more strict: It allows the use of text-to-speech on its online assessments, but only if the student is a nonreader who will never be able to read any words at any grade level throughout the student's lifetime. The student must also receive all printed material for every subject in an audio format, at all times.
Somewhat more widely used were read-aloud accommodations for test instructions and for math tests, the center found. PARCC and Smarter Balanced have both decided to leave that particular use of read-aloud accommodations up to the school-based special education team working with a given student.
In allowing the use of read-aloud accommodations, even with some caveats or limitations, the common-core-testing consortia are striking a different path from the National Assessment of Educational Progress, which examines a nationally representative sample of students at grades 4, 8, and 12. NAEP allows a wide range of accommodations, but does not allow passages to be read aloud to students. That restriction has led to large student exclusion rates in some states.
The National Center for Educational Outcomes has suggested that the common-core assessments offer an opportunity for states to develop a coherent, and in its view necessary, multistate policy.
Such a policy should take into account that some students may be able to draw inferences from a text, or answer questions based on its main idea, even if they struggle with translating letters into sounds, also known as decoding. Requiring those students to decode to answer a question based on text comprehension is an "artificial barrier," according to a report researchers at the center wrote as guidance for the testing consortia.
"Ensuring that common standards have addressed accessibility concerns does not mean lowering the standards. It does mean, for example, providing a way for students who cannot hear to demonstrate their 'listening' skills; for students who cannot see to demonstrate their 'viewing' skills; and for students who cannot decode to demonstrate their comprehension skills in reading," the report says. But it's not so easy to separate the tasks of reading comprehension and decoding, said literacy expert Timothy Shanahan, the chairman of the department of curriculum and instruction at the University of Illinois at Chicago.
"Part of the task of reading and learning to read is learning to get the words off the page while you think about them. Not having to get the words off the page gives a measurable advantage in most studies," he said.
In a blog post, Mr. Shanahan gave PARCC's read-aloud accommodation decision the "Lindsay Lohan Award for Bad Judgment," a cheeky reference to the troubled starlet. Smarter Balanced made a better move by limiting the read-aloud accommodation in the early grades, he said, but his preference would be not to use it at all.
"Part of my concern is I have absolutely no doubt there are going to be school districts and states where people are going to be shopping to get the highest scores they can get," he said. And in states where read-aloud accommodations are used liberally, "those kids are going to do better if they're read to than if they have to read it themselves."
Diane Cordry Golden, the policy coordinator for the Missouri Council of Administrators of Special Education and the project coordinator for the Association of Assistive Technology Act Programs, has a foot in the special education and the assistive-technology worlds. She criticizes views such as Mr. Shanahan's, saying they are akin to "telling a child who can walk with great difficulty that using a wheelchair is not as good. You need to walk, period," she said. "We can't give you technology to work around it; we've got to fix this thing that we see as not normal in you."
It's a legitimate concern that a read-aloud accommodation could be overused, Ms. Golden added, but "this summative test is not an appropriate mechanism to use to address those issues. If you have kids who are really failing readers because of poor instruction, that needs to be driven by something else other than calling out kids with disabilities."
The consortia are moving ahead with field-testing on a united front, though some state education leaders, such as in Colorado, a PARCC state, registered concerns about allowing the accommodation when the final vote was taken.
"Read-aloud is one of our major concerns. It's pretty fundamental to us," said Robert Hammond, Colorado's commissioner of education, at that meeting, held in June.
But both consortia say it's important to have some policy in place in time for the field-testing. If the accommodation is overused or badly deployed, the field tests will reveal those problems.
Mitchell D. Chester, the commissioner of education in Massachusetts, a state that does use a read-aloud accommodation on reading tests in limited circumstances, voted for the policy for the purpose of "getting it out into the bloodstream."
"I don't think there's a right or wrong answer to this," said Mr. Chester, the chairman of the PARCC governing board. "I think it's more a philosophical question."
Vol. 33, Issue 10, Pages s20, s21, s22Published in Print: October 30, 2013, as 'Read-Aloud' Assistance Puts Test-Makers on Spot