Skip over navigation

Journal Issue: Special Education for Students with Disabilities Volume 6 Number 1 Spring 1996

Identification and Assessment of Students with Disabilities
Daniel J. Reschly

Diagnosis, Classification,and Treatment

Elaborate legal requirements govern the procedures whereby a student is diagnosed as disabled and placed in special education. The process can be divided into several stages, each reflecting legally enforceable safeguards that are designed to ensure that students with disabilities are identified and provided special education and, at the same time, nondisabled students are protected from inappropriate placement. The stages are prereferral, referral, preplacement evaluation, eligibility determination, IEP development, determination of the placement, provision of services, annual evaluation of progress, and triennial reevaluation.

Progress from prereferral to the provision of services can be interrupted and halted at any one of the stages depending on the nature of the assessment information, professional judgment, and the decisions of parents. Informed parental consent is required prior to the initiation of the preplacement evaluation and again prior to the provision of services. It is at the preplacement and triennial reevaluation stages that decisions are made about eligibility for services under the IDEA. See the article by Martin and Martin in this journal issue for a discussion of the legal requirements and parental and student rights regarding special education evaluations.

Of all disability categories, mild learning disability may be the most difficult to diagnose. Yet, given the prevalence of this diagnosis, it is crucial that the process be examined. Eligibility for learning disability typically involves teacher or parent referral because of concerns about achievement lagging behind the child's apparent intelligence or measured IQ. The evaluation typically includes observation in the regular classroom, review of the child's educational history including past test scores, assessment with standardized tests of achievement and intellectual functioning, determination if there are any discrepancies between achievement and intellectual ability, and elimination of other possible causes of the learning problem (for example, sensory deficits).

In recent years increasing concern has been expressed regarding the dominance of standardized tests at the expense of assessment that is related to interventions in evaluations for learning disability and mild mental retardation. The administration of a comprehensive, individually administered IQ test and one or more standardized, individually administered achievement tests nearly always dominates the learning disability eligibility process. Such testing is virtually mandated by federal guidelines to establish a "severe discrepancy between achievement and intellectual ability."1


Problems with the current classification system were recognized at least 20 years ago in the large, federally-funded exceptional child classification project. Prevalent problems include stigma to the child, poor reliability for traditional categories, poor relation of categorization to treatment, obsolete assumptions still in use in treatment, and disproportionate representation of minority students.


The degree to which lifelong, permanent negative effects of classification (labeling) occur is disputed. Certainly, the more extreme claims made in the late 1960s, such as that labels create deviant behavior rather than vice-versa,20 are heard less often now. Nevertheless, the common names used for students with mild disabilities have negative connotations. An earlier, now classic, review21 reported that there is widespread misunderstanding of the meanings of traditional classifications by both professionals and the lay public;22 and the bearers of labels find the classification uncomfortable and, very often, objectionable.23 Concerns about the effects of classification on individuals have led to calls for the elimination of the common classification categories.24

Although this literature is complex, one conservative conclusion is that categorical classification should be used as sparingly as possible and, when used, should focus on skills rather than on presumed internal attributes of the individual. Current reforms that emphasize classification based on the specific skill deficits (low reading decoding skills) and the services needed (tutoring in phonological awareness) rather than presumed internal attributes may lessen the negative connotations.


Current diagnoses using traditional categories are frequently unreliable. Although it is virtually impossible for a student performing at the average level or above to be classified as learning disabled or mildly mentally retarded, differentiating between these categories or between these categories and other classifications such as slow learner, economically disadvantaged, and at risk for poor educational outcomes is often difficult. The reasons for this difficulty include (1) overlapping characteristics among students in these categories,25-27 (2) variations in teacher tolerance for student diversity (see the article by Hocutt in this journal issue), (3) differences in screening and placement practices among districts, and (4) variations in the quality of assessment measures used by professionals.28

Researchers19 have noted the diagnosis of dyslexia is not stable for children in the elementary grade levels. The instability from year to year further aggravates the reliability of the diagnosis of dyslexia, an important subcategory of learning disability.

Relation of Classification to Treatment

A disability category is useful to the degree that it is related to the determination of treatment, to treatment outcome, and/or to prevention. The information needed to determine whether or not a student is eligible to be classified as learning disabled, mildly mentally retarded, or seriously emotionally disturbed typically does not relate closely to treatment decisions regarding individual goals, objectives, monitoring of interventions, or evaluating outcomes. Furthermore, considerable evidence now suggests that the educational interventions provided to students in the different disability categories are more alike than different.23,29,30 Effective instructional programming utilizes the same principles and often the same procedures (intensive individual instruction, along with close monitoring and feedback) regardless of whether the student is classified as learning disabled, mildly mentally retarded, seriously emotionally disturbed, a slow learner, or educationally disadvantaged.30

Another criterion for usefulness is relation to prognosis or outcomes. The research has indicated that traditional categories do not have a demonstrable relationship to specific outcomes or to prognoses.30-32

Obsolete Assumption: Homogeneous, Segregated Groups

A subtle, but important, premise of the current categorical system is that students must be classified into categories so that homogeneous groups can be formed. The efficacy of programming by handicapping condition has been questioned since the 1960s and continues to be a subject of concern with regard to the current categorical system.23,29,31-33 Many education agencies and practitioners are moving away from the assumption that student services can be determined by category; it is time for the categorical system to reflect this change in practice.

Obsolete Assumption: Aptitude by Treatment Interaction

Perhaps the most widely accepted traditional assumption is that special intervention techniques, instructional methods, and instructional materials must be carefully matched to precisely diagnosed learning styles or processes. The underlying assumption in this matching process was that of an aptitude by treatment interaction (ATI).34 The ATI evidence, however, has been uniformly negative in special education applications using disability categories, modality preferences, learning styles, cognitive processing, or neuropsychologically "intact" areas.31,33,35-38 The process- or style-matching justification for the current categorical system has little empirical support.

Disproportionate Minority Placement

One of the most controversial aspects of the current system is the disproportionate placement of minority students in various categories of disability. Recent data regarding the participation of various groups of students in special education programs are summarized in Table 3. The data are subject to differing interpretations; however, the principal conclusions are (1) both African-American and Hispanic students are disproportionately represented in special education but in opposite directions, and (2) the disproportionately high number of African Americans in special education reflects the fact that more black students than white students are categorized as having mild mental retardation. Regardless of the actual proportions, there is widespread belief that special education has been used as a dumping ground for minority students.39

Commonly suggested causes of disproportionate minority representation in special education include (1) poverty, (2) discrimination or cultural bias in referral and assessment, and (3) unique factors related directly to race or ethnicity. Wagner's40 analyses implicated poverty as the principal reason African-American students are overrepresented in special education. A similar conclusion was published by Reschly41 in an analysis of a large sample of African- American and white students in Delaware who were classified as learning disabled. However, other studies have produced different results, and it cannot be assumed that poverty is the only, or primary, causative agent. Other factors, such as the increased prevalence of low birth weight among African Americans,42 should also be considered.

Positive Features of the Current Classification System

The current categorical system has served as (1) a rallying point for advocacy groups seeking support for programs, (2) the structure for passage of legislation, and (3) the basis for allocation of monies to establish educational services for students with disabilities. The monumental progress made over the past 30 years has occurred within the confines of the present categorical system. Efforts to reform the classification system need to provide plausible alternatives that ensure the continued social and political support for programs needed by students with disabilities.

Alternatives to the Current System

The overall goal of the special education disability classification system should be to enhance the quality of interventions and improve outcomes for children and youth with disabilities. At the same time, the categories used should be as free as possible of negative connotations, recognizing that no disability classification system will be totally free of negative connotations. This section recommends the development of systems organized around the supports and services needed by children and youth, with further designation, if needed, of the dimensions of behavior in which supports and services are provided.24,43

Dimensional, Not Typological

Classification systems should be based on dimensions of behavior (reading, social conduct, and the like) rather than on typologies of persons. Typologies involving dichotomies such as disabled-nondisabled, retarded-not retarded, and learning disabled-not learning disabled are never accurate reflections of the diversity of student aptitudes and achievement. As discussed earlier, students vary on broad continua by fine gradations. However, dichotomous decisions are imposed by the current classification system.

Current eligibility rules require educators to decide that virtually identical students have very different educational needs. These decisions are inaccurate. What is needed is a classification system that reflects the reality of student differences. A classification system based on broad dimensions with fine gradations would allow accurate description of the status of students without imposing false, either-or dichotomies.

In the meantime, there is some merit to the position taken by advocates for the learning disabled, calling for preservation of the full continuum of services. For the student diagnosed with mild learning disability, the school district, in combination with the parents, might be best advised to experiment with intense interventions (for example, temporary or long-term placement in a separate classroom), limited intervention (for example, small-group tutoring two or three times a week), or simply a wait-and-see approach (for example, no changes at school but intensive tutoring support from parents at home) based upon the family's preferences, the student's motivation, and the results of intervention. When the degree of disability can be measured but response to treatment cannot be predicted, the best choice may be to offer multiple treatment options.

Functional, Not Etiological

The current classification system is based primarily on etiology or presumed internal attributes of individuals. These etiological formulations are not useful in that they are not closely related to treatment.

For the vast majority of students now classified as mildly disabled, functional classification will mean emphasis on skills related to the school academic curriculum and to essential social competencies. Attempts to use functional classification criteria and programming have been successful and represent enormous promise for improving the current delivery system.44-47 This trend is by no means universal, nor even present in a majority of school districts. Important barriers in the forms of funding mechanisms and disability eligibility criteria exist in most states. However, these impediments have been placed under careful scrutiny in recent policy papers43 sponsored by the Federal Office of Special Education Programs.


All professionals and parents realize that students with disabilities are complex human beings with a wide range of assets and limitations. Unfortunately, the current classification system suggests that persons with disabilities are different from the norm on one or two salient dimensions such as intelligence or achievement. The focus on one or two dimensions rather than on the broad range of assets and limitations often leads to undesirable restrictions of programming to those dimensions. For example, although it is well known that a significant proportion of students with learning disability have difficulties with social skills, or that the adult adjustment of persons with mild mental retardation will be determined to a greater degree by social rather than by academic competencies, current educational programs often ignore the vital areas of social skills and social competencies.48

Reliable Technology

Over the past 20 years, a reliable technology has been developed for direct measurement of student behavior in natural settings.46,49-50 When an assessment reveals reliable and precise information about a student's deviations from the average on relevant dimensions, this information can be used in measuring the effectiveness of interventions (for example, assessment of current status in relation to target objectives, monitoring progress, and evaluating outcomes). Such detailed data on the degree of student variance from the norm could also be used in allocating services to students with the greatest needs, but it should be noted that this approach may encourage the assignment of limited resources primarily to students with the more severe behavioral problems, giving a lower priority to early intervention for students whose problems are not yet extreme.

Knowledge Based on Effective Intervention

Clearly, there is a body of knowledge related to the effectiveness of instructional interventions. Classification systems that focus on functional dimensions of behavior will facilitate the application of that knowledge base. In contrast, a classification system that focuses on presumed etiology, or on factors such as underlying neuropsychological processes or learning modalities that have no relationship to treatment outcomes, interferes with the provision of effective treatment.