LD summit: important issues for the field of learning disabilities.

Abstract. This article summarizes nine White Papers that were commissioned by the Office of Special Education Programs (OSEP) and presented at the Learning Disabilities Summit -- Building a Foundation for the Future, August 2001. The purpose of the article is to inform the membership of the Council for Learning Disabilities (CLD) of the issues identified by authors of the White Papers that promise to influence research, policy, and practice.

**********

On August 27 and 28, 2001, the Office of Special Education Programs (OSEP) of the U.S. Department of Education held a Learning Disabilities (LD) Summit in Washington, DC. Secretary Rod Paige addressed more than 200 educators, researchers, advocates, and policy makers during the plenary session. Dr. Paige emphasized the importance of developing a "genuine science" for dealing with learning disabilities and stated that IDEA reauthorization will be guided by four principles: accountability for results, local control and flexibility, expanded parent options, and linking research-based strategies to the classroom. Recently appointed Assistant Secretary Robert Pasternak noted that "the [Bush] administration has a commitment to this review [of IDEA] to determine whether changes are needed to evaluate children diagnosed with learning disabilities."

The LD Summit -- Building a Foundation for the Future was part of the national LD Initiative undertaken by OSEP. During the mid-nineties, the National Joint Committee on Learning Disabilities (1) (NJCLD) requested that OSEP examine federal regulations relating to identification of children and adolescents with LD. With reauthorization of the Individuals with Disabilities Education Act imminent, OSEP deferred examining (nonstatutory) LD identification procedures.

In an effort to fulfill its commitment to NJCLD's request, OSEP began its national LD Initiative in spring of 2000 after deciding that the first step was to clearly articulate the current LD research base. The following summer, OSEP convened a work group of 21 researchers, service providers, policy makers, and parents to identify White Paper topics, paper authors, and paper respondents. This work resulted in the development of nine White Papers and more than 30 response papers, which were presented by the authors during the 2001 summit. This information will help guide policy makers during the reauthorization of IDEA '97 to ensure that students with LD are properly identified and provided with a free appropriate education.

With the call for full funding of IDEA, LD identification promises to be a big issue during the reauthorization process as this disability category currently accounts for more than half of all students served by special education. In addition, there have been calls by some to overhaul special education and to radically rethink how we identify and serve students with disabilities. For example, consider these statements made by the editors of the recently published Fordham Foundation's Rethinking Special Education for a New Century:

   Putting it bluntly, special education is broken for too many children.
   Think of it as "program at risk." Special education now has far too many
   categories -- particularly in the "LD' area -- and is too vague about which
   children need this assistance. Rather than today's "one-size-fits-all" IDEA
   mandates and procedures, policymakers should consider creating two or
   possibly three categories of students within the special needs population.
   Learning disabilities pose issues of their own. Exactly what gives students
   with specific learning problems an entitlement to greater education
   resources than their peers who simply are slow learners and/or struggling
   for other reasons? (Finn, Rotherham, & Hokanson, 2001, pp. 336-345)

It is clear that special education, and the learning disabilities field in particular, faces significant challenges as reauthorization of IDEA 97 begins next year. The purpose of this article is to provide an overview of each summit White Paper, (2) which identified issues facing the LD field, and to report efforts CLD will undertake to respond to these issues in order to inform policy and practice.

WHITE PAPERS

Early Childhood/Early Identification

In their paper, "Early Identification and Intervention for Young Children with Reading/Learning Disabilities," Jenkins and O'Connor (2001) provide an overview of the reading difficulties students with reading/learning disabilities exhibit and a summary of the research findings on early identification and intervention. The authors identify the abilities that skillful readers possess to comprehend text, including phonological awareness (PA), letter-sound correspondence, decoding, automatic word recognition, fluency, and language comprehension (Adams, 1990). Research findings support the notion that students with reading disabilities manifest difficulties with these skills and exhibit marked differences in these skills from their peers. Thus, there is a need for early identification and intervention to help reduce the incidence or severity of reading disabilities.

Early identification research is rooted in concurrent and predictive studies that focus on preliteracy skills and reading ability. Findings have emerged to support the contribution of skills such as PA and letter knowledge to selected reading skills. However, the authors discuss the issue of underprediction and overprediction associated with assessing the relationship between phonological abilities and later reading skill. Acknowledging that underprediction misses students who could possibly benefit from early intervention and that overprediction taxes resources and contributes to misinterpretation of intervention validity, the authors identify efforts underway to address the prediction issue.

The authors provide information about and critical issues associated with early intervention emphasizing PA and letter-sound knowledge. The predictive value of PA and the relationship of PA to understanding the alphabetic principle have influenced the recommendation to implement phonemic awareness instruction with young children. In particular, researchers have sought to understand the benefits of PA as a part of early preliteracy instruction and the contribution of phonemic awareness instruction to decoding instruction for at-risk students. Questions have been raised regarding how PA can best be taught and what levels of intensity and achievement benchmarks are necessary. Research findings suggest the need to examine more closely the issues of duration and intensity of instruction for students who are not making adequate progress.

Recognizing that students with reading disabilities exhibit difficulties with decoding and fluency, the authors identify issues that have been the focus of research, including delivery of instruction (i.e., more-and less-explicit instruction), instructional value of emphasizing phonemes or rimes (i.e., phonograms, word families), and the effects of decoding instruction on word-level reading ability (National Reading Panel Report, 2000).

Critical intervention questions remain unanswered. Research is needed on the role of decodable text in word-level reading, the criterion for mastery that will promote automatic word identification and comprehension, and the practices that best promote fluent reading, including decontextualized word-level reading. Finally, research is needed to better determine specialized instructional practices for youngsters who do not demonstrate sufficient learning despite our best treatment efforts. The authors conclude that prevention and intervention of reading/learning disabilities is desirable and that much is yet to be learned.

Classification Approach

In their paper, "Classification of Learning Disabilities: An Evidenced-Based Evaluation," Fletcher, Lyon, Barnes, Stuebing, Francis, Olson, Shaywitz, and Shaywitz (2001) examine the validity and reliability of current procedures used to classify students as having learning disabilities (LD). Largely based on the results of NICHD-sponsored research focusing on early reading acquisition, they conclude that little difference exists between poor readers with no demonstrated discrepancy between ability and achievement (i.e., students who are low achievers) and poor readers who demonstrate a discrepancy (i.e., students who have LD).

Fletcher et al. suggest that ability-achievement discrepancy (in their view, the only inclusionary element of the federal LD definition) is not a phenomenon restricted to students with LD; underachievement among students with LD is to be expected. Further, based upon results of the Connecticut Longitudinal Study (Francis, Shaywitz, Stuebing, Shaywitz, & Fletcher, 1996), there appear to be no differences in growth rates of level of reading ability for groups of low-achieving (LA) students and children with learning disabilities (LD).

When considering the effect of IQ on reading performance, Fletcher and his colleagues report that lower and higher ability children respond similarly to explicit phonics instruction used for both prevention and remediation of reading failure. Thus, response to treatment fails to differentiate between groups of LA and LD students. Fletcher et al. propose that alternative definitions of LD be developed in order to (a) identify domain-specific types of LD (e.g., dyslexia) and (b) lead toward use of interventions that specifically target the LD subtype (e.g., instruction in phonological processing skills).

Based upon their review of the research, Fletcher et al. conclude that students with LD may demonstrate difficulties in and within one or more academic areas, thus supporting the heterogeneity classification hypothesis. Results of current research have identified at least two types of reading disability (i.e., decoding and comprehension) and at least one type of math disability (i.e., calculation).

According to Fletcher et al., it is appropriate to exclude students with mental retardation and sensory disorders from LD classification systems since these students' treatment needs differ from those of students with LD. However, they question the validity of excluding from consideration of learning disabilities students with emotional and behavioral disabilities. Similarly, Fletcher and his colleagues question the validity of excluding students who are socially, culturally, or economically disadvantaged. They maintain that serving some students in special education (i.e., students with LD) and others in Title 1 (i.e., LA group) programs creates a potentially inequitable two-tier service delivery system.

An underlying theme of this paper is the importance of prevention. Fletcher et al. criticize the current LD identification process as allowing children to fail until they receive services, emphasizing the need for alternative definitions of LD to be treatment-oriented. In summary, Fletcher et al. maintain that differentiation between LA and LD groups lacks validity when prognosis, response to intervention, neurobiological factors, and cognitive, behavioral, and academic correlates are considered.

Historical Perspective

Hallahan and Mercer (2001), in their paper, "Learning Disabilities: Historical Perspectives," offer a historical overview of learning disabilities by considering LD within five time periods -- underscoring the longstanding recognition of the concept of learning disabilities.

The European Foundation Period (1800 to 1920) was marked by basic scientific research focusing on brain function and dysfunction. Pioneers in the field included Goldstein, Werner, Strauss, Lehtinen, Cruickshank, and Kephart.

Research focusing on the attention, perceptual, and perceptual-motor disabilities of adults with brain injuries was conducted by Goldstein, Werner, Strauss, Lehtinen, Cruickshank, and Kephart during the U.S. Foundation Period (1920 to 1960). Also, the effect of these disabilities upon children's reading and language was investigated by Orton and Fernald, who developed instructional intervention programs still in use today.

During the Emergent Period (1960 to 1975), the term learning disabilities was first used by Samuel Kirk, and parent and professional organizations devoted to LD were formed. Interventions focusing on remediation of psychological processing and perceptual problems were predominant. In addition, the work of federally sponsored task forces was incorporated into legislation, including the Children with Specific Learning Disabilities Act of 1969.

The Solidification Period (1975 to 1985) began with the passage of the Education for All Handicapped Children Act (i.e., Public Law 94-142), which included LD as one of the disability categories. In 1975, the 1968 definition of the National Advisory Committee on Handicapped Children was incorporated into Public Law 94-142. In 1977, federal regulations established criteria for identification of students with LD, including the ability-achievement discrepancy criterion. In addition, applied research conducted at universities, including five federally funded LD research institutes, resulted in the adoption of empirically validated educational procedures such as cognitive strategies instruction, cognitive behavior modification, direct instruction, and the abandonment of process training.

Hallahan and Mercer call the current era the Turbulent Period (1985 to 2000). Although positive strides have been made in identifying effective methods to teach children to read and to use strategies to enhance learning, dissension exists in the LD field. According to Hallahan and Mercer, the following four issues are fueling tension and discord:

1. Students have been overidentified/misidentified as LD.

2. Minority students are disproportionately represented in the LD category.

3. The ability-achievement discrepancy lacks validity when identifying presence of a learning disabililty.

4. Full inclusion denies some students with LD an appropriate education.

Hallahan and Mercer suggest that these issues become polarized because individuals adopt modernist or postmodernist perspectives. Modernists regard LD as a real condition that transcends social context, but one that can be treated or accommodated through the provision of special education. Postmodernists, on the other hand, regard LD as a socially constructed phenomenon that is an indefensible disability category used to segregate students based upon race or class. Rather than focus on treating the child, postmodernists advocate changing social constructions, which in their view restrict individuals with disabilities. As Hallahan and Mercer note, "These two views are contending vigorously as the configuration and mission of special education continue to evolve" (p. 7).

Approaches to Decision Making

In their paper, "Learning Disabilities as Operationally Defined by Schools," MacMillan and Siperstein (2001) contrast school-based and researcher identification practices that result in two different groups of students: school-identified (SI LD) and research-identified (RI LD) students with learning disabilities. School personnel tend to use individual and collective subjective judgment and local norms in order to serve students they perceive are most in need of services. Researchers, on the other hand, identify students with LD by applying criteria included in the federal definition.

MacMillan and Siperstein suggest that the "false-positive children with LD" whom schools identify need to receive services consistent with their etiologies and needs. They further recommend that researchers and practitioners agree on how LD is identified, emphasizing that current practices based on a "wait and fail" precedent, such as meeting an ability-achievement discrepancy criterion, are harmful to students.

The key implication of school identification practices is revealed in MacMillan and Siperstein's statement, "as such, LD is currently operationally defined in the public schools as absolute low achievement, not necessarily discrepancy from aptitude and not necessarily excluding cases [more accurately accounted for by the exclusionary criteria]" (p. 7). In other words, schools use LD as a category of failure, and to the degree schools apply objective standardized identification criteria, they creatively employ them to get the results they desire. Put even more straightforwardly, schools want LD to be something it officially is not, or at least provide benefits of the LD label to a group that does not resemble research-identified children. However, research-based definitions and identification criteria are not conclusive enough to discount school practices.

As a means of rectifying this situation, school-based practitioners and researchers should collaborate to determine what LD really is and to convincingly distinguish whether and how students with true LD require separate services. This requires reconsidering both the definition and how it is operationalized (Kavale & Forness, 2000). At present a schism exists between research and practice. Because RI students fail to resemble SI students, research fails to adequately inform practice. Finally, MacMillan and Siperstein warn that in order to serve "traditional" LD students (RI LD) by preserving the principle of appropriate individualized education, LD professionals need to join forces with those who advocate that "nontraditional" LD students (SI LD) also receive effective instruction.

Discrepancy Model

In his paper, "Discrepancy Models in the Identification of Learning Disability," Kavale (2001) reviews issues around the use of discrepancy models in the identification of LD. Even though discrepancy has been the primary criterion for defining LD since 1976, the federal definition does not mention discrepancy specifically. Kavale maintains that discrepancy is a useful and valid tool in the identification of LD by demonstrating the difference between ability and achievement even though quantifying the discrepancy and application of the findings remains a problem.

Specifically, determining the best means for calculating the discrepancy between ability and achievement is problematic. Although the standard score formula or a regression score table is most often used to determine whether a discrepancy exists, some identification teams use the difference between grade placement and grade achievement level or a formula based on grade-equivalent scores. Consequently, there is potential variability and inaccuracy associated with discrepancy models. For example, students who are identified as LD in one state or system may not qualify in another state or system.

Another issue addressed in this paper is the increased number of students identified as LD, partly because discrepancy between ability and achievement is often ignored. Eligibility teams run the gamut, from strictly adhering to the discrepancy rule for identification, to exclusively relying on results of a discrepancy formula, to disregarding ability-achievement discrepancy entirely. This lack of consistency may lead to subjective identification practices resulting in large numbers of low achievers being identified as having LD. Students often are identified as LD solely based on low achievement, with many also exhibiting lower than average intelligence. According to Kavale, this practice endangers the existence of LD as a category.

Kavale views the procedures schools use to determine a discrepancy between ability and achievement as an important issue in itself. Problems with IQ tests, and with what they measure, have been acknowledged. Limitations of achievement tests have been noted, and guidelines recommending specific achievement measures currently are unavailable. Best practice suggests that a combination of standardized, norm-referenced, and informal tools be used to assess achievement. However, there is lack of consistency in practice.

As Kavale states, "discrepancy documents the presence of underachievement, not LD" (p. 7). The discrepancy should be only one part of the identification process, not the end all. Kavale suggests that the process he and Forness (2000) developed be followed, whereby discrepancy is used as the first of a five-step identification process. In summary, Kavale regards ability-achievement discrepancy as a valid and reliable concept and practice and suggests that discrepancy is a useful component in the LD identification process.

Alternative Responses to Intervention

In his paper, "Responsiveness to Intervention: An Alternative Approach to the Identification of Learning Disabilities," Gresham (2001) identifies two major reasons for considering alternatives to the current discrepancy formula for identification of students with LD. First, the information yielded by the current discrepancy formula does not contribute to useful planning, implementing, and evaluating of instructional interventions. Second, there is no direct link between assessment procedures and interventions. Gresham describes the current use of discrepancy as focusing on discrepancy between a student's ability and achievement levels, as measured by standardized assessments. As an alternative, he suggests a discrepancy-based approach to identification that ties assessment to interventions: a pre- and post-intervention assessment that depicts how well a student responds to interventions. Using this approach, a student is identified as having LD only if his or her performance does not improve satisfactorily when a validated intervention is implemented with integrity. Three components of Gresham's recommendation are key: (a) a student's performance must be measured, (b) an intervention that is likely to improve performance must be selected, and (c) the intervention must be implemented appropriately (i.e., treatment validity).

Essential to determining the effectiveness of an intervention is the assessment used to monitor a student's performance during instruction. To that end, Gresham maintains that curriculum-based measurement may be used to make eligibility decisions (Stecker & Fuchs, 2000). Another data-based model that may be used to monitor instructional effectiveness, and that also includes examination of contextual factors (i.e., antecedent and consequent events), is applied behavior analysis. Moreover, the assessment tools must be accompanied by interventions that are validated and include the best predictors of a student's performance in the area of concern. For example, for students who have problems with reading, measurement criteria such as reading accuracy, reading rate, and reading comprehension would be used concurrent with interventions such as direct instruction and strategy training models (Swanson, 1999).

Gresham notes that some researchers suggest a dual-discrepancy model for identifying students with learning disabilities. With a dual-discrepancy model, not only would a student perform below the level of classroom peers but his or her rate of learning would be below that of peers. Subsequent eligibility for special education services would continue to measure instructional effectiveness to ensure improved student growth. Regardless of whether or not a student is found eligible for special education services, the measurement and intervention process continues as evidence of effectiveness of and responsiveness to interventions.

Gresham acknowledges that using a responsiveness to intervention discrepancy model raises issues regarding which intervention to use, how extensive and intensive the intervention needs to be, ensuring the integrity of how the intervention is implemented, and using good judgments or decision rules for defining satisfactory instruction. Moreover, a cost-benefit analysis is recommended to determine the usefulness and costs of responsiveness to intervention.

Processing Deficit Models

Torgesen (2001) describes what he considers the best case that can currently be made in support of the concept and practice of direct diagnosis of learning disabilities by assessment of intrinsic processing weaknesses in his paper, "Empirical and Theoretical Support for Direct Diagnosis of Learning Disabilities by Assessment of Intrinsic Processing Weaknesses." His scholarly treatise serves an important function to the field, as it accurately describes the current state of affairs on this issue. In a word, we are currently unable to diagnose learning disabilities directly and reliably by assessment of intrinsic processing weaknesses. Although direct diagnosis of intrinsic processing weaknesses has been suggested as a promising replacement over ability-achievement discrepancy and diagnosis by exclusion, Torgesen suggests that we lack reliable and valid assessment approaches to make this a viable alternative today.

Although Torgesen questions our ability to directly assess intrinsic processing weaknesses, he notes the considerable progress the field of LD has made during the past two decades. Specifically, great strides have been made in understanding the role of intermediate skills such as phonological processing in the development of basic reading and its manifestations of learning disabilities. Torgesen acknowledges, however, that in the case of phonologically based reading disabilities, the tasks used to predict and diagnose reading disabilities are not direct measures of the processing weaknesses fundamental to the disorder. Torgesen recommends using intermediate predictor variables such as phonological processing in lieu of intrinsic process measures to diagnose LD. It is difficult to know for certain whether performance problems on psychological tests reflect intrinsic processing limitations or whether performance is limited by deficits in acquired knowledge. Finally, he cautions that our knowledge of the effects of intermediate skills upon listening, language, and mathematics is extremely limited.

Rather than support assessment of intrinsic processing weakness to diagnose LD, Torgesen recommends a two-tier model that combines assessment of predictor variables (i.e., Phonemic awareness, understanding of letter-sound correspondence) with monitoring of children's response to intervention. Diagnosis of LD would not occur until the third grade or later, after a child has failed to respond to interventions carefully designed to prevent reading problems. This two-tier model shares the advantages of a processing foundation, while it also provides specifically tailored instruction to students with similar learning problems. By focusing instruction on areas of greatest need, Torgesen maintains that this model is superior to discrepancy-based approaches.

Torgesen clarifies the record regarding using intrinsic processing weaknesses in the direct diagnosis of learning disabilities. However desirable it may be to be able to diagnose LD using a processing approach, it is not possible at present; we must wait until we are able to demonstrate convincingly the reliability and validity of a diagnostic model that requires direct assessment of intrinsic processing weaknesses. The advances in the field cited by Torgesen provide evidence that process assessment warrants further investigation.

Clinical Judgments

Wise and Snyder (2001), in "Judgments in Identifying and Teaching Children with Language-Based Reading Difficulties," describe how expert clinicians use and go beyond scores derived from standardized tests to develop the most appropriate educational programs for children with reading difficulties. Reading receives emphasis because of the high prevalence of reading problems in children referred for special education, as well as its salient importance.

Wise and Snyder focus on two types of reading disabilities: one based on underlying core deficits in phonological (or speech-sound based) processing, leading to primary difficulties in word reading; another based on poor language comprehension, in which children struggle with main ideas, summaries, and inferences whether they read or listen to stories, despite normal phonological processing skills. After reviewing the research on identifying and instructing students with speech-sound based reading disabilities, the authors conclude that expert teachers and clinicians must assess foundation-level skills, ensure automaticity with speeded practice, and provide numerous opportunities for children to apply the skills in context at instructional levels, while at the same time supporting comprehension in individualized programs and encouraging independent reading and writing away from the clinical setting to ensure transfer.

Stating that specific comprehension problems have not been studied closely in the United States, Wise and Snyder draw from research in the United Kingdom on identifying and instructing children with specific comprehension problems. They conclude that promising but limited findings indicate that focused intervention using strategies to address skills related to semantic memory seems to improve reading comprehension performance. Citing some U.S. research, they point to interventions that used self-questioning, small-group instruction, and extensive practice at levels that ensure success as important for improving comprehension.

Wise and Snyder discuss the role clinical judgment plays in recognizing children with reading disabilities, making the point that referring a child to special education is a signal that the teacher has used clinical judgment in differentially evaluating students' classroom behaviors. They believe referral is a sign that teachers have reached their level of tolerance of individual differences and are no longer optimistic about their ability to effectively instruct the referred student. The authors provide information on how to clinically recognize children with speech-sound based reading disabilities and children with specific poor comprehension as well as information about screening processes for reading disabilities and their research underpinnings.

According to Wise and Snyder, teachers need a strong knowledge base in language and all aspects of reading to make the kind of judgments and research-based modifications recommended in their paper. In order for teachers to be able to implement clinical judgments in modifying instruction to meet student needs, Wise and Snyder urge schools of education to require more courses in language, reading, and individualizing instruction for all elementary school and special education teachers, and school districts to provide inservice training for all teachers to help support and enrich their knowledge.

Wise and Snyder conclude by stating that the best instruction for students with LD supplements the prescriptions derived from formal assessments with clinical judgment of students' responsiveness to teaching and ongoing assessments of student progress and classroom performance.

Is LD Just a Fancy Term for Low Achievement?

A continuing controversy in the field of learning disabilities (LD) is if (and if so, how) students identified as having LD differ from nondisabled low-achieving students. With extremely large and increasing numbers of underperforming students being identified as having LD, it is essential that we determine if the LD construct is a valid and useful designation in the design and delivery of remedial and special education services. In their article, "Is `Learning Disabilities' Just a Fancy Term for Low Achievement: A Meta-Analysis of Reading Differences Between Low Achievers With and Without the Label," Fuchs, Fuchs, Mathes, Lipsey, and Roberts (2001) note that if students identified as having LD cannot be distinguished from their nonlabeled low-achieving classmates, it would seem reasonable to abolish the disability category and to focus efforts and resources on raising achievement levels of all low-performing students. However, before denying special services to the large numbers of students who receive special education as a result of the LD construct, empirical efforts must seek to document the overlap between LD and low achievement (LA).

To determine the degree of overlap (or difference) between LD and LA, Fuchs et al. identified and quantitatively synthesized all published and unpublished studies on the reading achievement of students with LD and LA nondisabled students. The researchers selected the reading domain because (a) reading is the focus of the majority of studies comparing LA and students with and without an LD diagnosis, (b) reading deficiencies are a major cause of a LD designation, and (c) reading difficulties affect most aspects of academic performance.

After reviewing and integrating the results of 73 studies, Fuchs et al. found an effect size of .6 of a standard deviation, indicating that the reading scores of 73% of the LA population fell above the average reading scores of the population of students with LD. Independent factors explaining the greatest amount of the effect size variance were test format, grade levels of the students with LD, and the type of measurement used in the identification of LD. Specifically, comparative studies that employed timed tests of reading, involved participants with LD from higher grade levels, and used objective measurement in the identification of LD had larger effect sizes than studies employing untimed tests of performance, involved students at lower grades, and made considerable use of teacher and team judgments of LD.

The Fuchs et al. data indicate that students with LD have more severe differences in reading performance than students who are LA. These differences tend to be most dramatic when (a) timed rather untimed tests are used, (b) students are in higher grades and experience accumulating deficiencies in reading, and (c) objective test scores rather than human judgment are used in the diagnosis of LD. Without evidence of a qualitatively different set of learner characteristics, one that would require a unique educational program, it may well be that differences among students with LD and LA are more a matter of degree rather than of kind. Similar to other syndromes that are based on degree of difference from a range of acceptable levels, such as hypertension and obesity, LD may be a disability that is defined simply by extreme low achievement. This extreme low achievement, particularly in timed situations and in higher grades that require skilled comprehension, will likely require a specialized educational intervention.

SUMMARY

In previous sections of this article, each of the nine White Papers presented during the 2001 LD Summit was summarized. These papers underscore the need to examine our practices regarding classification and identification of students with LD. White Paper authors challenge assumptions regarding the federal definition and its accompanying identification criteria, such as requiring discrepancy between ability and achievement as well as excluding certain groups of students. Several themes are evident within and across the White Papers.

Early Intervention/Prevention

Many White Paper authors stress the need for early intervention efforts and recognize the need to adopt effective instructional practice to prevent educational failure of large groups of students. Specifically, the current practice of documenting ability-achievement discrepancy before LD services are provided was characterized by several authors as a "wait and fail" approach that must be abandoned.

LD as Low Achievement

Several White paper authors contend that we cannot differentiate LD from "garden variety" low achievement. For example, Fletcher et al. (2001) conclude that little difference exists between poor readers with no demonstrated ability-achievement discrepancy and poor readers who demonstrate a discrepancy. Several authors maintain that "unexpected underachievement" among students with LD is indeed to be expected. Some authors regard students with LD as the lowest of low-performing students, leading Fuchs et al. (2001) to suggest that extremely low academic achievement be used as the primary criterion to identify LD.

Discrepancy Alternatives

No consensus was reached on the validity of discrepancy as a construct essential to LD. Even those who support the construct were dissatisfied with the way it is currently operationalized (Kavale, 2001). However, few viable alternatives were proposed to take the place of ability-achievement discrepancy when identifying LD. Because intrinsic processing deficits, rather than underachievement, are thought by some to be the underlying cause of LD, direct assessment of these processes has intuitive appeal but is beyond our capability at this time (Torgesen, 2001). Support for clinical judgment in the identification process was equivocal (MacMillan & Siperstein, 2001; Wise & Snyder, 2001).

Response to Treatment

Many White Paper authors acknowledge that there remains a group of students who fail to respond to even the best intervention approaches, and that a substantial number of children resist early intervention efforts. For example, Torgesen (2000) estimates that approximately 2 to 6 percent of children remain poor readers despite receiving prevention interventions. These "treatment resisters" may constitute the group of students who require LD services. Several authors recommend conceptualizing LD as responsiveness to intervention rather than ability-achievement discrepancy (Fuchs et al., 2001; Gresham, 2001).

Research, Policy, and Practice Incongruence 

Several White Papers note that students identified as having LD by school personnel fail to meet federal definition/identification criteria. While researchers tend to regard students with LD as those meeting these criteria, school decision-makers tend to rely on local norms and subjective judgments to identify students most in need of academic assistance. Many schools identify students who have low IQ scores and who fail to demonstrate a discrepancy between ability and achievement as LD. School personnel want to provide services to low-performing students, regardless of state education code and authoritative definitions (MacMillan & Siperstein, 2001).

FUTURE DIRECTIONS FOR CLD

CLD was well represented at the LD Summit, with many past and present members of the Board of Trustees in attendance, including the authors of this article. In addition, CLD is currently participating in an NJCLD roundtable process facilitated by NCLD and funded by OSEP. Representatives of NJCLD organizations met in October to identify common ground responses to issues raised at the LD Summit, identify issues requiring further discussion, and provide OSEP with an NJCLD roundtable final report in April 2002. Five broad categories will be used to organize issues addressed (and not addressed) during the summit: eligibility criteria, identification process, intervention, professional development, and nature of LD. Each NJCLD member organization will develop papers for each of these five areas.

As part of this overall effort, CLD roundtables were held during the 23rd International Conference on Learning Disabilities recently held in Charlotte, North Carolina. Issues identified during the roundtables will be incorporated into the CLD Issues Papers, which will present our organization's views of the most critical issues facing LD and their impact on research, policy, and practice. The content of our papers will form the basis for development of the NJCLD final consensus report, which will be delivered to OSEP for consideration during the reauthorization process.

FOOTNOTES

(1) In addition to CLD, NJCLD member organizations include the American Speech-Language-Hearing-Association, Division for Learning Disabilities, International Reading Association, National Association of School Psychologists, Association for Higher Education and Disability, Division for Children's Communication Development, International Dyslexia Association, Learning Disabilities Association of America, and National Center for Learning Disabilities.

(2) Executive summaries of White Papers, tapes and transcripts of presentations may be accessed at www.air.org/ldsummit or www.ed.gov/offices/OSERS/OSEP. Summaries in this article are based upon executive summaries rather than full-length White Papers. White Papers and response papers will be released by OSEP as a set of proceedings in the near future.

REFERENCES

Adams, M.J. (1990). Beginning to read. Cambridge, MA: MIT Press.

Finn, C. E., Rotherham, A. J., & Hokanson, C. R., Jr. (2001). Conclusions and principles for reform. In C. E. Finn, A. J. Rotherham, & C. R. Hokanson, Jr. (Eds.), Rethinking special education for a new century (pp. 335-347). Washington, DC: Thomas B. Fordham Foundation.

Fletcher, J. M., Lyon, G. R., Barnes, M., Stuebing, K. K., Francis, D. J., Olson R. K., Shaywitz, S. E., & Shaywitz, B. A. (2001, October 1). Classification of learning disabilities: An evidence-based evaluation. Paper presented at the 2001 Learning Disabilities Summit -- Building a Foundation for the Future. Executive summary retrieved October 1, 2001, from www.air.org/ldsummit.

Francis, D. J., Shaywitz, S. E., Stuebing, K. K., Shaywitz, B. A., & Fletcher, J. M. (1996). Developmental lag versus deficit models of reading disability: A longitudinal individual growth curves analysis. Journal of Educational Psychology, 88, 3-17.

Fuchs, D., Fuchs, L. S., Mathes, P. G., Lipsey, M. W., & Roberts, P. H. (2001, October 1). Is "learning disabilities" just a fancy term for low achievement: A meta-analysis of reading differences between low achievers with and without the label. Paper presented at the 2001 Learning Disabilities Summit -- Building a Foundation for the Future. Executive summary retrieved October 1, 2001, from www.air.org/ldsummit.

Gresham, F. (2001, October 1). Responsiveness to intervention: An alternative approach to the identification of learning disabilities. Paper presented at the 2001 Learning Disabilities Summit -- Building a Foundation for the Future. Executive summary retrieved October 1, 2001, from www.air.org/ldsummit.

Hallahan, D. P., & Mercer, C. D. (2001, October 1). Learning disabilities: Historical perspectives. Paper presented at the 2001 Learning Disabilities Summit -- Building a Foundation for the Future. Executive summary retrieved October 1, 2001, from www.air.org/ldsummit.

Jenkins, J., & O'Connor, R. (2001, October 1). Early identification and intervention for young children with reading/learning disabilities. Paper presented at the 2001 Learning Disabilities Summit -- Building a Foundation for the Future. Executive summary retrieved October 1, 2001, from www.air.org/ldsummit.

Kavale, K. A. (2001, October 1). Discrepancy models in the identification of learning disability. Paper presented at the 2001 Learning Disabilities Summit -- Building a Foundation for the Future. Executive summary retrieved October 1, 2001, from www.air.org/ldsummit.

Kavale, K. A., & Forness, S. R. (2000). What definitions of learning disability say and don't say: A critical analysis. Journal of Learning Disabilities, 33, 239-256.

MacMillan, D. L., & Siperstein, G. N. (2001, October 1). Learning disabilities as operationally defined by schools. Paper presented at the 2001 Learning Disabilities Summit -- Building a Foundation for the Future. Executive summary retrieved October 1, 2001, from www.air.org/ldsummit.

National Reading Panel Report. (2000). Report of the national reading panel. Teaching children to read. Washington, DC: National Institutes of Health.

Stecker, P. M., & Fuchs, L. S. (2000). Effecting superior achievement using curriculum-based measurement: The importance of individual progress monitoring. Learning Disabilities Research & Practice, 15, 128-134.

Swanson, H. L. (1999). Instructional components that predict treatment outcomes for students with learning disabilities: Support for a combined strategy and direct instruction model. Learning Disabilities Research & Practice, 14, 129-140.

Swanson, H.L. (2000). Issues facing the field of learning disabilities. Learning Disability Quarterly, 23, 37-49.

Torgesen, J. J. (2001, October 1). Empirical and theoretical support for direct diagnosis of learning disabilities by assessment of intrinsic processing weaknesses. Paper presented at the 2001 Learning Disabilities Summit -- Building a Foundation for the Future. Executive summary retrieved October 1, 2001, from www.air.org/ldsummit.

Torgesen, J.S. (2000). Individual differences in response to early interventions in reading: The lingering problem of treatment resisters. Learning Disabilities Research & Practice, 15(1), 55-64.

Wise, B. W., & Snyder, L. (2001, October 1). Judgments in identifying and teaching children with language-based reading difficulties. Paper presented at the 2001 Learning Disabilities Summit -- Building a Foundation for the Future. Executive summary retrieved October 1, 2001, from www.air.org/ldsummit.

Linda K. Elksnin. Ph.D., is Professor of Education, The Citadel.
Diane Pedrotty Bryant, Ph.D., is Associate Professor,
University of Texas-Austin.
Debi Gartland, Ph.D., is Professor of Special Education,
Towson University.
Margaret King-Sears, Ph.D., is Associate Professor,
Johns Hopkins University.
Michael S. Rosenberg, Ph.D., is Professor of Education,
Johns Hopkins University.
David Scanlon, Ph.D., is Assistant Professor of
Special Education, Boston College.
Roberta Strosnider, Ph.D., is Associate Professor of
Special Education, Towson University.
Rich Wilson, Ph.D., is Professor of Special Education, Bowling
Green State University.