Once upon a time, there was a town whose playground was at the edge of a cliff. Every so often a child would fall off the cliff. Finally, the town council decided that something should be done about the serious injuries to children. After much discussion, however, the council was deadlocked. Some council members wanted to put a fence at the top of the cliff, but others wanted to put an ambulance at the bottom.
In this parable, the idea of putting an ambulance at the bottom of the cliff clearly is foolish on many levels. Waiting for children to be injured and only then providing them with help is cruel and inhuman if the damage can be prevented. Further, it is needlessly expensive; an ambulance costs far more than a fence.
Yet longstanding policies in special education, especially for children with learning disabilities, are very much like this ill-considered idea. Schools generally provide pretty good programs in kindergarten, 1st grade, and beyond, but they know with certainty that a number of children will fall by the wayside. In particular, a certain number of children of normal intelligence will fail to learn to read. After a while, these children are very likely to be retained, assigned to long-term remedial services, or labeled as having specific learning disabilities and provided with special education services.
By the time these services are rendered, most of the children will already have realized that they have failed at their most important task—learning to read. Accordingly, they likely will have lost much of their earlier motivation, enthusiasm, and positive expectations. Schools will be paying for years—in special education and in remedial instruction costs—for failing to ensure that students succeed in the early grades.
Neverstreaming: Does It Work?
Today, most children with learning disabilities are mainstreamed for much of their school day. Mainstreaming is better than self-contained placement, but it is far from ideal. Mainstreamed students with academic deficiencies often are poorly accepted by their peers, struggle with academic content, and develop low self-esteem (Bear et al. 1991).
Obviously, students fare better when they succeed the first time they are taught, thereby avoiding both special education and mainstreaming. We call this neverstreaming: implementing prevention and early intervention programs powerful enough to ensure that virtually every child is successful in the first place (Slavin et al. 1991).
Few people would deny that this is a nice policy in concept. The question is, would it be practical with real kids in real schools? Evidence is accumulating that it is in fact possible to ensure the success of almost all children in the early elementary grades—at least in reading. And this has profound implications for special education for children with reading disabilities. Let's look at that evidence.
Exhibit 1: Success for All
Perhaps the strongest evidence for the feasibility of neverstreaming comes from research in our own Success for All program (Slavin et al. 1996), a comprehensive approach to restructuring elementary schools. The Johns Hopkins University's Center for Research on the Education of Students Placed at Risk (CRESPAR) is implementing the program in more than 300 schools in 70 districts in 24 states. The focus is prevention and intensive early intervention for children in preschool through 6th grade.
As preventive measures, we provide research-based instruction in reading, writing, and language arts. The programs emphasize cooperative learning and maintain a balance among phonics, children's literature, creative writing, and home reading. We back up these programs by offering (1) intensive professional development, (2) a full-time building facilitator to help teachers continue to improve their instructional strategies, (3) a curriculum-based assessment program to monitor student progress and identify children in need of additional help, and (4) a strong parent involvement program.
Even the best instructional programs cannot ensure success for every child. For this reason, Success for All schools also provide one-to-one tutoring for 1st graders who are struggling in reading. The tutors typically are certified teachers. Their goal is to see that the students never become remedial readers. To do this, they provide students with instruction that is closely tied to classroom reading instruction and teach them metacognitive skills—for example, how to ask themselves whether what they read makes sense and relates to what they've been taught.
A family support team in each school also provides intensive early intervention.Team members get parents involved and give them strategies for helping their own children. They also develop programs to improve attendance, resolve behavior problems, and work with local agencies to see that children have eyeglasses, hearing aids, health services, or other needed assistance.
Research on Success for All (Slavin et al. 1992, 1994, 1996; Madden et al. 1993) has shown that the program has consistently improved children's reading skills, as measured by both individually administered reading tests and standardized assessments. Students in nine school districts across the United States have average scores three months ahead of those of matched control students by the end of 1st grade, and more than a year ahead by the end of 5th grade. The effects are particularly dramatic for students who are most at risk, those in the lowest quarter of their grades.
These findings have direct relevance to special education. A longitudinal study in very high-poverty Baltimore schools found that of 3rd graders participating in the program, including those in special education, only 2 percent were performing two years below grade level, the usual criterion for identifying reading disabilities. In the control groups, by contrast, 9 percent of 3rd graders were performing this poorly (Slavin et al. 1992). Overall, special education placements were cut in half.
In a study conducted in Ft. Wayne, Indiana, Smith and colleagues found that special education referrals in grades K–3 were more than three times higher in control schools than in Success for All schools. In another study, across four districts, Smith and colleagues also found that 1st grade special education students in Success for All schools had substantially higher achievement than 1st graders in control schools (Smith et al. 1994).
Success for All is usually funded by reallocation of Title I funds, supplemented on occasion by special education funds.Very few schools have funding beyond what they would have had without the program. This means that even without extra resources, schools can substantially reduce reading failure. With additional resources to pay for more tutors, they can further reduce the number of children failing to meet adequate reading standards (see Slavin et al. 1992).
Exhibit 2: Reading Recovery
Reading Recovery is a 1st grade tutoring program being used successfully in thousands of U.S. schools. It was developed in New Zealand by Marie Clay (1985), and researched and disseminated in the United States by Gay Su Pinnell and her colleagues at Ohio State University (1988).
The program provides 30 minutes of daily, one-to-one tutoring to 1st graders who score poorly on a battery of diagnostic tests. Tutors are certified teachers who complete an extensive professional development program.
As with Success for All, research on Reading Recovery shows that the vast majority of children can be well on the way to reading effectively by the end of 1st grade (Pinnell et al. 1994). Further, Reading Recovery students are more likely than matched comparison students to stay out of special education (Lyons 1989).
Exhibit 3: Prevention of Learning Disabilities
As the name suggests, Prevention of Learning Disabilities is designed to keep children from ever needing special education services for learning disabilities (Silver and Hagin 1990). Like Reading Recovery, this program provides one-to-one tutoring to at-risk 1st graders. The tutoring, however, focuses on general perceptual skills as well as reading. Studies of the program have found that 1st graders have substantially higher achievement than do similar control groups (Silver and Hagin 1990).
Exhibit 4: Early Childhood Interventions
The Carolina Abecedarian Project (Campbell and Ramey 1994) is the best example of an intensive early intervention program. Research on this program has found that following children through the critical first five years of their lives has strong and lasting effects.
Reading Recovery and Prevention of Learning Disabilities start with children in 1st grade, and Success for All starts with 4- and 5-year-olds. Yet much of children's cognitive development has already taken place by age 4 (Carnegie Corporation 1994). This project offers further evidence that if young children growing up in poverty are effectively stimulated and their parents are helped to create a healthy home environment, they are more likely to perform well in school and to stay out of special education.
Exhibit 5: Family Support and Integrated Services
Two national programs designed to improve school-home collaboration and services for children in the early grades are Comer's (1988) School Development Program and Zigler's Schools of the 21st Century (Zigler et al. 1992). These programs aim to head off the causes of school failure that have nothing to do with children's cognitive capabilities: conflicts between parents' and schools' values and expectations; erratic attendance; or the need for eyeglasses, hearing aids, or more adequate nutrition, for example. Family support and integrated service programs can solve many of these problems.
The Verdict Is In
Up to now, no one program has been able to ensure that all children are reading well enough to stay out of special education. However, the programs described here have come close, and with additional research and experience are sure to come even closer.
Moreover, imagine what would happen if all children in need of early intervention participated in multiple programs as intensive and comprehensive as these, from preschool through the elementary grades. The number of children still having reading problems would almost certainly be a fraction of what it is today.
A useful and appropriate debate is going on about special education versus inclusion for children with more serious, low-incidence disabilities. For children at risk of learning disabilities, however, neither of these alternatives is the answer. Instead, we need to focus on prevention and early intervention. If we know how to ensure that virtually every child will become a skillful, strategic, and enthusiastic reader, then it is criminal to let children fall behind and only then provide assistance. Neverstreaming, not mainstreaming or special education, should be the goal for all children who are at risk.
