The handmade flashcards were not helping my nephew Clayton. My sister Linda confided: "He's not reading. We practice, but he can't remember the words the next time. He gets frustrated."
Although it seemed overwhelming, Clayton's problem was fairly simple. "If Clayton is reading the word cat," Linda explained, "he just says the letters c, a, t. He doesn't recognize the word." Clayton wasn't connecting the letters to the sounds they represent. Children often are taught the names of letters first, which can make it hard to learn how they're pronounced. For these kids, the letter c has no relationship to the sound k in cat.
Compounding the problem, Clayton's kindergarten teacher was giving him word lists to memorize, failing to recognize that he didn't know the basic letter sounds. She kept sending home new lists even though he hadn't learned the words on the previous ones. It's not surprising that Linda and Clayton were frustrated.
I was worried for Clayton because I know what happens to kids when they don't learn to read. Comprehensive research by the National Institute for Child Health and Human Development shows that children who cannot identify word sounds in kindergarten often cannot read by third grade. If Clayton failed to learn the relationship between letters and sounds in kindergarten, chances are he would be assigned to special education by fourth grade, which would spell his doom in the public school system. He probably would never become a proficient reader. Despite attending a solidly middle-class school, rated 7 out of 10 by the state of California, Clayton could easily end up as yet another child labeled "learning disabled" because his school failed to teach him how to read.
This winter Congress is scheduled to reauthorize the Individuals With Disabilities in Education Act (IDEA), which dispenses $60 billion a year to school districts around the country. While there's no question that IDEA has provided legal protections and services for students with handicaps, it has also created perverse incentives that encourage schools to call kids disabled as a way of attracting more funding and masking instructional failures. Instead of restructuring the program to mitigate these unintended consequences, Congress is set to simply throw more money at the problem.
Disability As an Excuse
Nearly 12 percent of American students in kindergarten through 12th grade are assigned to the special education system. Children with severe disabilities, such as mental retardation, autism, blindness, and deafness, account for only a tenth of these students. The remaining 90 percent are described as suffering from conditions that are less obvious and harder to verify objectively, such as specific learning disability (SLD), speech and language delays, mild mental retardation, and emotional disorders. SLD is the most common label, accounting for more than half of all students covered by IDEA. SLD diagnoses, which have risen by 34 percent since 1991, are the main factor contributing to the dramatic increase in special education enrollments since 1976.
In a recent Education Week commentary, Manhattan Institute education analyst Jay Greene observes that the SLD category "has more than tripled from 1.8% of the student population in 1976?7 to 6.0% in 1998?9. All other categories of special education combined…have actually declined from 6.5% to 5.8% of the student population during the same period." Greene sees these trends as cause for skepticism about the validity of SLD designations. "If a general increase were truly underway in the proportion of students with learning problems," he writes, "then it should be evident in more than just one category of special education."
Federal law defines SLD as "a disorder in one or more of the basic psychological processes involved in understanding or in using language, spoken or written, which disorder may manifest itself in imperfect ability to listen, think, speak, read, write, spell, or do mathematical calculations." To prevent overuse of the label, federal regulations stipulate that it be limited to students who show a "severe discrepancy" between their achievement in one or more subject areas and their intelligence, usually as measured by an IQ test. For example, a child who scores lower on a standardized reading test than on an IQ test might be classified as having a reading disability.
Even with these criteria, an SLD diagnosis remains subjective. In addition to the federal standard, there are 50 different state definitions of learning disability, and the methods used to determine intelligence vary widely. University of Minnesota education researchers James Ysseldyke and Bob Algozzine estimate that more than 80 percent of all schoolchildren in the United States could qualify as learning disabled under one definition or another. In a 1986 study, UCLA education psychologist Esther Sinclair and her colleagues applied five different formulas to a sample of 137 children. Those classified as learning disabled ranged from 4 percent to 28 percent.
Andrew J. Coulson sums it up neatly in his book Market Education: The Unknown History. "In the world of public schooling," he writes, "SLD diagnosis is often reduced to a devastatingly simple formula: if a child is smart but cannot read or do math, he is disabled." A consensus report published by the National Institute of Child Health and Human Development in 2001 concluded that it's impossible to clearly distinguish between an SLD in reading and low achievement: "Dyslexic children simply represent the lower portion of the continuum of reading capabilities."
A 2002 report from the President's Commission on Special Education estimated that 80 percent of students who receive an SLD diagnosis–two out of five special education students–are assigned to the program "simply because they haven't learned how to read." In a similar vein, an in-depth analysis in Rethinking Special Education for a New Century, a 2001 report published by the Fordham Foundation and the Progressive Policy Institute, estimates that nearly 2 million children would not have been classified as learning disabled if the public schools they attended had provided proper, rigorous, and early reading instruction.
In this connection, it's instructive to compare IDEA to Title I, which funds "remedial" reading and math instruction for children from poor families. Any student who qualifies for the federal free lunch program is eligible for Title I services. Although the government distinguishes between special education, intended for students described as disabled, and remedial education, intended for students presumed to be at a disadvantage because of their economic background, the same sort of intensive instruction seems to work equally well for poor readers in both groups. In fact, schools often pool money from both programs to pay for one general intervention, such as reading resource labs.
The SLD label is increasingly popular not because it suggests a particular pedagogical approach but because it brings schools extra money. The incentive to identify students as disabled is especially strong in schools with large numbers of low-income students. Such schools can obtain funding under Title I as well as IDEA, double counting each low achiever. "In essence," write Wade Horn and Douglas Tynan, "low-income, low-achieving students can be 'twofers' when it comes to maximizing procurement of federal and state funds."
It is commonly asserted that special education puts a financial strain on schools. Yet during the last four decades per pupil spending has increased from $2,360 to $7,086 in inflation-adjusted dollars, while student outcomes have been flat. "Whatever the causes for this productivity crisis in education (spending more without improving outcomes)," the Manhattan Institute's Jay Greene notes, "it is not reasonable to blame special education for consuming extra dollars or burdening schools with more difficult to educate students." Even as they shift more and more students into special education, schools have more money for general education than ever before. "Schools are classifying more normal but low-achieving students as learning-disabled using vague criteria," Greene writes. "Schools get more money for these special-education kids but don't spend much to 'treat' them."
This trend is especially troubling when one considers a child's dismal chances of learning to read through special education. The longer students remain in special education, the lower their reading ability when compared to that of other poor readers. As Louise Spear-Swerling and Robert J. Sternberg explain in their 1998 book Off Track: When Poor Readers Become "Learning Disabled," "Poor readers in special education may be particularly likely to suffer decreases in practice, to benefit less from instructional interaction with a teacher, to engage in unmotivating instructional activities, and to draw maladaptive conclusions about what reading is." Similarly, a 1989 study by education researchers Richard Allington and Anne McGill-Franzen found that poor readers in special education programs received less instructional time in reading than did regular classroom students or Title I students. A 2000 survey of 500 special education teachers by the Council for Exceptional Children found that most reported devoting less than one hour a week to one-on-one time with students.
Unlike special education, early intervention with intensive instruction appears to reduce the number of children who have reading difficulties later in life. The research suggests that when children like my nephew Clayton are taught the basic phonological skills necessary for reading, they can avoid a disability label altogether. The experience with early intervention programs that emphasize phonemes (basic units of speech) indicates that the rate of truly intractable reading problems is close to the rate of other serious disabilities. In five recent studies, when kids with poor phonological skills were given intensive instruction in phonemes and phonics, the expected incidence of learning disabilities, originally 12 percent to 18 percent, was reduced to around 1.5 percent.
"The emphasis on prevention begs the question of what constitutes a disability," write reading expert Reid Lyon and his colleagues in the Rethinking Special Education report. "If the role of inadequate instruction is taken seriously, and more aggressive attempts are made to teach all children to read, the meaning of disability could change in the future. In this scenario, the actual diagnosis of LD could be reserved for children whose reading or other academic problems are severe and intractable."
Full Funding of a Bad IDEA
Lyon argues that complex assessments and disability determinations should be replaced by a system offering intensive instruction to all children who score below the 25th percentile in reading achievement. Whatever the eligibility criteria, it's vital that funding be tied to performance. In this respect, policy makers can learn something from child welfare reform.
Foster care funding is usually based on how many days children remain in the system; the longer they stay, the more revenue they generate. The unintended consequence is that kids languish in foster care, neither reunited with their natural parents nor adopted by new parents. Some innovative states, such as Kansas and Michigan, have tied foster care payments to the speed with which agencies find permanent placements for children. Agencies that move children into permanent family arrangements more quickly receive more money. Similarly, a better approach to special education would reward states that lower their disability rates through intensive early intervention.
Special education voucher programs would also help correct the perverse incentives created by IDEA. First, vouchers would allow parents to find the school environment that best fits their children's circumstances. Second, vouchers would discourage schools from overidentifying learning disabilities: Better to teach students to read in the first place than lose their per pupil revenue altogether.
Unfortunately, it's unlikely that Congress will rethink special education. It appears that "full funding" of IDEA–defined as covering 40 percent of the extra cost to educate disabled children–is a done deal. The Senate IDEA reauthorization plan calls for a $2.5 billion annual increase in appropriations, resulting in full funding in six years. The House Republican plan calls for full funding of IDEA in 10 years, including annual increases of just over $1 billion through fiscal year 2007. Large, predictable funding increases can be expected to encourage further expansion of special education as schools strive to maximize their budgets.
My nephew Clayton already brings his school Title I money, based on my sister's income and his eligibility for the free lunch program. The good news is that my sister has prevented Clayton's school from turning him into a "twofer" and assigning him to special education. Linda took responsibility for teaching Clayton the relationship between letters and sounds, and by the end of the school year he had jumped from a 1 to a 3 (on a scale of 1 to 4) in his kindergarten reading classification. It's the sort of success that could be far more common if schools focused on teaching kids to read rather than diagnosing their disabilities.