Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Unlabeled But Still Entitled: Toward More Effective Redemption

by Gaea Leinhardt, William Bickel & Allan Pallay - 1982

Compensatory education students and mildly handicapped special education students would be better served by a single system meeting the needs of both groups. Student classification methods, historical, political, and pedagogical reasons for separating the programs, and alternatives to the existing arrangement are discussed. (Source: ERIC)

Twenty-five percent of the public elementary school students in the United States are currently considered unable to learn to read, write, and compute under “normal” or regular classroom instructional conditions.1 These students have not been ignored. They are, in fact, bombarded with state and federal legislation to support innovative remedial treatments. Although there are many local variations, the two treatment programs that have been most often promulgated are generally known as special education and compensatory education. These two categories can be thought of as subsystems within the educational environment designed to allocate instructional and social resources to students who are, for a variety of reasons, not experiencing adequate progress in the schools. Under one or the other (or, in some cases, both) of these categories, students are screened, evaluated, and classified for treatment. This discussion concerns the nature of these classifications, the historical and political reasons for their existence, their pedagogical justification, and alternatives to existing schemes.

Compensatory education is a system of educational and social service programs that have been designed to help students who are both poor and not achieving well in school. Depending on the percentage of poor children in a specific school, an eligible child may or may not receive extra educational treatment. The system of programs is largely federally funded with some state support in larger states. It is administered through state and local educational agencies. Funds are used primarily to purchase instructional aids and additional equipment.

Special education is a system of educational programs that have been designed to help students who do not learn well under ordinary school conditions. The most significant difference between compensatory and special education is that special education requires the child to be diagnosed as having a definable and inherent disability. Once identified, the student is entitled by law to a special, individualized educational program to help him reach his potential. The special services are funded by local, state, and federal sources. These funds are used primarily for hiring trained teachers and aides, purchasing additional equipment, and instruction in smaller self-contained classes or part-time resource rooms.

Both special education and compensatory education classify students who are either doing poorly in school or are at great risk of doing so. In the case of compensatory education, the classification is based on the income of the student’s family. Theoretically, a student who is not poor should not be served by compensatory education programs. In the case of special education, the classification is based on a set of cognitive and affective characteristics of the student. The classifications are presumably related to types of instruction that are suited to children in the particular category. As will be pointed out later, neither knowledge of the social class of a child nor which of the three categories the child is in helps in making educational decisions.

The central position of this article is that the educational needs of compensatory education students and mildly handicapped special education students could be more adequately met by a single system serving both groups of children.2 In such a system, children would not be distinguished as compensatory education students or special education students. They would be recognized as students who need additional educational service. We support this position by arguing that there is no educational utility for the separation. That is, evidence will be presented suggesting that the treatments that are effective for both groups of students are similar. Furthermore, we will take the position that a merged system would help reduce some of the negative aspects of these programs and magnify the positive ones.3

Our purpose is to provide the basis for discussion of what we believe to be a major question: Should these elaborate and separate classification and treatment systems remain as separate entities, or should they be combined? The evidence for a decision to combine the systems is limited to a specific instructional area, the attainment of basic skills in reading and math. In this domain, we will present evidence that suggests that what is succeeding in one program area will, in all likelihood, work in other areas as well. We freely admit that the evidence is, in some instances, tentative, and that in any case cognitive growth represents only one of a number of goals that these two systems have. Nevertheless, the similarity in successful treatments in this area casts serious doubt on the overall instructional rationale that supports separate systems.

There are, of course, important historical reasons for the separate development of the systems that go beyond the instructional goals of these programs. These historical factors have led to systems that have somewhat different structures, operating procedures, and goals. These historical factors have also led to a set of political, professional, and philosophical forces that help maintain the separation and could hinder the development and operation of a merged system. How these factors might affect the promulgation of a merged system will be discussed in the concluding section of the article.

This article is divided into four major sections. First, similarities and differences between special education and compensatory education classifications, origins, goals, practices, and consequences are described. Second, the rationales for these classifications are examined. Third, one specific rationale, the pedagogical one, is critically analyzed in detail and tested for its soundness in the face of new evidence on the effectiveness of current treatment practices. Finally, an alternative to the current classification approach is discussed (along with political constraints that may hinder implementation) that may possibly increase the effectiveness of the remedial programs while decreasing the potential negative effects of current practices.



As universal education became a reality in the United States in the post-World War II era, it became painfully clear that equal educational opportunity was more complex than providing students with identical instructional resources. Concern for children who, as a group, came from poor families and who were performing one standard deviation or more below the rest of American students grew steadily. The basic concern for this group was that children who started out behind their fellows would remain so and fall continuously further behind. These students were likely candidates to drop out at a point when they were still unable, like their parents before them, to compete successfully in society. This problem (inability to learn in school leading to inability to perform in work) was identified as the vicious cycle of poverty.

A simultaneous attack on many sides of this cycle was launched as part of Lyndon Johnson’s War on Poverty. A key front in this war was to be the classroom. The nation would find many ways to break the back of ignorance—before regular school, during school, after school—with specially trained teachers and new and innovative techniques.

The hope was that students who did poorly in school only because their parents had little money could be rescued relatively easily. If a student was successful in school, this would dramatically improve his chances for success in life. The key elements that influenced and still influence attitudes and treatments in compensatory education programs are: (1) children from poor families are more likely than average to do poorly in school; (2) their poor performance is due to social and economic conditions in the home and community that affected the individual rather than to something organic within them; and (3) even though these conditions are not treated, progress can be made in treating the child’s academic problems.

Federally supported compensatory education has its functional bases in two legislative acts, the Economic Opportunity Act (EOA) passed in 1964 (and amendments to it),4 and the Elementary and Secondary Education Act (ESEA) passed in 1965.5 These acts established Head Start, Title I, and Follow Through programs. Compensatory education has its theoretical roots in two developments (one social, one educational) that occurred during the decade prior to the enactment of the federal legislation.

The choice of education as the means to improve conditions of the poor was based on a number of factors. Some were political (e.g., education of the young was a relatively benign and socially acceptable type of federal intervention), and some were based on the personal convictions of the presidents involved (Johnson, a former teacher in a one-room schoolhouse, was particularly adamant about the power of education to affect life chances). However, of interest here is the theoretical basis for early intervention that had evolved in the social sciences during the 1950s and early 1960s which provided an educational justification for compensatory education.

Just as economists, social planners, and others had rediscovered poverty in the early 1960s, psychologists and other social scientists in the 1950s had “rediscovered” the power of the environment to influence human behavior. Zigler and Anderson6 credit the work of two men, Benjamin Bloom7 and J. McVicker Hunt,8 with turning over the traditional view that immutable genetic capacity in the individual dominated development. Hunt emphasized the power of the mother on intellectual growth. Bloom emphasized the critical need for early intervention. These views when combined with those of anthropologist Oscar Lewis, who described a “culture of poverty” passed on from generation to generation,9provided a theoretical rationale for compensatory education. Effective social intervention in the early years was both possible and necessary for the sound development of the individual.


Three federal initiatives, Head Start, Title I, and Follow Through, form the core of compensatory educational programming over the past one and one half decades. Among the three programs, well over $3.5 billion is expended annually, with Title I at roughly $3 billion the largest single component. A review of the legislative history of these programs is an important prerequisite to understanding their goals.

Head Start

Head Start, a preschool instructional initiative, was conceived in 1964 and first implemented during the summer of 1965. Head Start was not the subject of a specific act of Congress. Rather, the first Head Start summer programs were implemented under the legislative authority of the Economic Opportunity Act of 1964.10 The programs were to be administered by the Office of Economic Opportunity (OEO), which had been created by the act to wage the war on poverty. While P.L. 88-452 did not make specific reference to Head Start, it did direct OEO to pay particular attention to the needs of the young.

The location of Head Start evolved from the Johnson administration’s concern for identifying ways to overcome local hostility to the massive federal intervention envisioned in OEO. Head Start, within OEO and under the direct administration of local Community Action Program (CAP) agencies, was in part viewed as providing such a mechanism. Sargent Shriver, OEO’s first director, stated that

faced with local and congressional hostility [to OEO] I felt that certain National Emphasis programs [i.e., programs originating in Washington but administered at the local level] one of them being Head Start, could ameliorate some of this hostility . . . by establishing certain national programs that many communities would consider desirable.11

The location of Head Start in OEO and with CAP agencies had an important influence on the original goals of the program.12

Head Start was originally conceived as, and still remains, a broad social service program aimed at low-income preschool-age children. The fundamental goals of the program were contained in recommendations of an OEO planning committee instituted in 1965 by Sargent Shriver and chaired by Robert Cooke, then pediatrician-in-chief of Johns Hopkins Hospital. The recommendations of this committee outlined seven major goals of the program for children: improving physical health and abilities, aiding social-emotional development, improving mental processes and skills, establishing patterns of success, improving family relationships, developing social responsibility, increasing a sense of worth and dignity.13

The target population for Head Start is preschool-age children, from three years old to school entry. A child who qualifies for Head Start gains access if the community has a Head Start program available, and if he meets poverty requirements.

Title I

Title I is a massive program providing direct support to school districts that are heavily impacted by high numbers of poor and low-achieving students. The program currently reaches 87 percent of the school districts in the country.14 The federal authority for Title I rests with the Elementary and Secondary Education Act of 1965.15 Title I, one of five titles originally in ESEA, provides financial assistance to local educational agencies for “the education of children of low-income families.“16 The other titles provide resources for libraries and instructional materials, supplementary educational centers, educational research and training, and strengthening state departments of education.

Two major issues (federal aid to education and federal aid to church-related schools), one nested within the other, surrounded the passage of ESEA and influenced its characteristics. The largest issue concerned the bill’s unprecedented infusion of massive amounts of federal dollars into education. During the post-World War II period, there had been several attempts in Congress to provide direct aid to schools. For the most part these efforts failed due to a general mistrust of the federal government’s involvement in local affairs. A more specific aspect of this opposition to federal aid related to the implications these resources would have for private, especially parochial, schools.

Maximum flexibility was given to local educational leaders. No specific educational treatments were mandated, although provision for some local evaluation was included in the law. The bill was clearly a resource distribution bill, rather than an educational bill per se. The law simply specified that “in recognition of the special needs of children of low-income families and the impact that concentrations of low-income families have on the ability of local educational agencies to support adequate educational monograms . . . the United States . . . [should] provide financial assistance.“17 Like those of the Head Start program, Title I goals reflected more clearly the political climate of the time than any specific educational theories or goals.

The target population of Title I is defined essentially by income and achievement requirements. Grants are distributed to states and then to districts on the basis of income levels. Within districts, schools are identified on the basis of poverty indices as well. Within the school, students achieving below grade level (exactly how far behind depends on age) on some standardized measure receive services. States have tended to vary how they distribute their Title I allocations. Some states have attempted to serve as many eligible students as possible, thus reducing per pupil expenditures. Other states focused their resources on smaller numbers of students. A 1968 survey documented a per pupil Title I expenditure range of $50 to $410.18Title I currently serves 15.1 percent of the total elementary population.19

Follow Through

On January 10, 1967, President Johnson in a state of the union address suggested that $120 million be allocated through amendments to the Economic Opportunity Act (1964) for a Fellow Through program that would extend the Head Start preschool efforts through grades K-3. Perhaps the most significant factor that stimulated the Johnson administration’s proposal was the issuance of an evaluation report on Head Start the previous year.20 This report asserted that Head Start classes were effective (based on interviews with teachers and parents, and tests of children) and speculated that the positive impact of Head Start was lost in the nonsupportive environment of public schools. For this latter point, the lack of positive difference between former Head Start participants and non-Head Start students was used.

The responsibility for the apparent failure to maintain presumed Head Start advantages (which were not clearly proven) was placed squarely at the door of the nonsupportive public schools. The response of the federal government was to extend the effort to the primary grades. The program as envisioned by Johnson was, like Head Start, a broad social-action program designed to reach thousands of students.

Planning for Follow Through as a large service-oriented program continued through 1967. However, by the time the amendments to the Economic Opportunity Act were passed (December 1967), an increasing congressional hostility toward the Community Action Program was reflected in a $200 million cut for OEO. This had an immediate impact on Follow Through plans. The Follow Through budget was slashed to $15 million. The conception of Follow Through as a national service program was clearly in doubt.21 The new identity became something known as planned variation where experimentation to find effective treatments rather than direct service delivery was emphasized.22 The nature and specific goals of planned variation will be described in the subsequent section on goals of compensatory education programs.

Follow Through was targeted to pick up where Head Start left off. Thus, Follow Through is in place from kindergarten through third or fourth grades. It was “designed to provide comprehensive services . . . [to] aid in the continued development of the children to their full potential.“23 The revision of Follow Through from a service program to one of planned variation also meant that far fewer children (e.g., 75,000 in 1976-1977) would be served. The original guidelines emphasized prior Head Start experience as an eligibility requirement, but this was not a meaningful restriction once the planned variation approach was developed. A child who qualifies for Follow Through gains access if the district is one that is part of the network, the school he attends is participating, and the principal has placed the child in the Follow Through classroom.


Unlike compensatory education, which arose from the attempt to ameliorate the problem of poverty, the special education of the mentally handicapped arose from the societal response to the deviancy and dependency caused by mental handicaps. Before the nineteenth century, mental handicaps were dealt with by the family and the local community. Publicly supported treatment on a large scale began in the twentieth century. The multiple theories explaining the roots of mental handicaps present in the nineteenth century coalesced into the widely accepted medical model early in twentieth.

The medical conceptualization of mild mental deficiency had the following rather Aristotelean flavor. The problem is seen as a personal disability or deficiency inherent in the individual. Ideally, a careful diagnosis and classification based on the presumed cause is done. Treatments are tailored to the illness that has been diagnosed. These treatments are administered by professional experts who have obtained special knowledge about the causes and cures of the problem. This model is consistent with the separate classification and treatment of the Learning Disabled (LD), Educable Mentally Retarded (EMR), and Socially and Emotionally Disturbed (SED) that have evolved over this century. That is to say, the problems of each group theoretically are unique and should be differentially treated. It should be noted that many of the features of the medical model of special education have come under attack in recent years.24

The medical model of special education provides an interesting contrast to the compensatory education conceptualization of poor school performance, which stresses the social origins of the problem. The school problems of the compensatory education child are not seen as arising from a personal disability but from a deprivation of school-relevant experience. The compensatory education teacher is not seen as a therapist curing a personal illness but as a teacher-social reformer who is providing extra educational resources to compensate for the lack of such resources at home. These fundamentally different philosophies partially explain why compensatory education and special education have been and continue to remain separate systems.


The first two major classifications of special education students were EMR and SED. Later, a variety of classifications, which now fall within the LD category, were added. Undoubtedly at the time that EMR and SED categories were formulated, children who were classified as mildly handicapped included those who are now served by compensatory programs. Special classes were provided for students who were thought to be in the various classifications. It was (and still is) difficult to categorize students into the EMR, LD, and SED classifications, and often there were not enough resources for separate classes. However, it was considered ideal if separation could be achieved and special classes provided.25 The first self-contained class for retarded children was inaugurated in Rhode Island in 1896. The numbers of children educated throughout the country in EMR classes rose steadily until the mid-1970s. By 1932 there were 75,099 students, by 1948 there were 82,471, by 1972 there were 785,000, and currently there are 711,373.26

The first special class for children with mild behavior problems was established in New Haven in 1871. In contrast with the EMR classes, classes for SED children grew very slowly before World War II. By 1948 only 15,340 children were served in these types of programs. In the 1950s there was an increased interest in the education of emotionally disturbed children. Many more special classes were established. Twenty-eight thousand children were in special day classes by 1958. The pace of development accelerated again in the 1960s and 1970s, and by 1979 there were over 330,000 students in classes for the emotionally disturbed.

The history of the identification and treatment of children with learning disabilities differs from that of EMR and SED in the following ways: First, it is a relatively newly recognized problem; second, it had an apparently low incidence prior to the 1950s; third, in contrast with EMR and SED, which are school-driven categories, LD came into being as a result of parental organizations and pressure groups.27

In 1963, local associations of parents of LD children met for a national convention. This convention produced two very important results: the coining of the term “learning disabilities,” uniting the various terms describing deficits other than EMR and SED, and the forming of the Association for Children with Learning Disabilities. A few years later, the Division for Children with Learning Disabilities was organized as a professional division within the Council for Exceptional Children. Since then, the growth has been explosive. In 1968 the number of children served was 120,000; by 1979, 1,270,551.28

While the pressure groups that lobbied for increased public services for LD and EMR children tended to be composed of separate individuals, the social-political climate that resulted in favorable legislation and court cases was the same. Unlike EMR, however, there has been no counter pressure from minorities or others against the LD label. This is because the consequences of being LD are not perceived as irreversible or all-pervasive. It is possible that the negative pressure placed on EMR has been in part responsible for the unexpected tenfold growth of LD, and that the decline in the number of EMR children served represents a shift from EMR to LD classifications.


This period can be broken into two parts: the postwar era through 1970 and the time from 1970 to the present. The expansion during the postwar era occurred as a result of local and state initiatives. Federal involvement tended to be restricted to modest support for personnel training and research (e.g., P.L. 85-926, passed in 1958). Reynolds and Birch describe two basic forces that influenced the growth of special education services in the public schools during this period: the formation of large, effective parent groups, and state initiatives mandating instruction.29

The 1970s encompassed a second period of special education expansion. Three features mark this most recent period, two of which contributed to the expansion and one of which slowed it. First, parent pressure groups increasingly turned to the courts as a means of changing school policy (e.g., PARC v. Pennsylvania, 1971).30Second, the emphasis on change moved generally from state and local arenas to the federal level (e.g., the Office of Civil Rights’ implementation of P.L. 93-l12, the Rehabilitation Act of 1973,31and, most significantly, the passage of P.L. 94-142, the Education for All Handicapped Children Act of 197532). Third, there was an emergence of pressure to limit the expansion from minority parents concerned about the disproportionate representation of minorities in certain special education categories (e.g., EMR and SED).


The broad goal of special education is the same as that of regular education. It is to provide each student with the skills necessary to realize his full potential for living and working.33 Since children who are thought to be in need of special education are often judged to have reduced potential, the concrete academic goals for special education students are often lower than those for students who do not receive special education. In addition, special educational programs often emphasize vocational and social skills more than do regular educational programs.34

Special education goals vary by program to some extent. The goals of special education are broader and more “normal” for the LD and SED students than for the EMR students. An SED child can overcome his problem and proceed in school in the same way as a “normal” child. An LD child can use his intellectual strengths in one area to compensate for weaknessess in another area and attain academic success. On the other hand, an EMR child is judged to have a general intellectual deficit and hence has no intellectual strengths that can be used to compensate for apparent weaknesses. Furthermore, it is generally believed (although not required) that a retarded student does not recover from his condition. If a retarded student should begin to perform at an average or above-average level, it would probably be assumed that the child was initially misdiagnosed. A diagnosis of retardation implies a very strong statement about potential. The students who have this label are the ones who are most likely to be placed in vocationally oriented programs early in their educational career. Besides the vocational emphasis, EMR programs often sacrifice time in academics to provide extra time for work on social and emotional issues.

The aim of compensatory education is to break the cycle of poverty by increasing the skills of the young generation of poor. The aim of special education is to help each student reach his full potential. While these goals sound somewhat different, in practice each program expends most of its time and money on the same thing: increasing the academic, and to a lesser extent the affective, skills of students who are not doing well in school.


Access to special education is almost baroque in its complexity. A child who is not doing well in school or who may be expected to do poorly or who just seems different may be nominated for evaluation by a teacher aware of the student’s performance. That person fills out a referral and gives it to the principal, regular education supervisor, or guidance counselor, who then contacts a psychologist (or a child study team) for evaluation. The psychologist or evaluation group contacts the parents for permission to evaluate. The child is tested. P.L. 94-142 requires that a variety of measures be used. However, current research indicates that there is still heavy emphasis on standardized achievement and IQ tests, especially in EMR placements.35 Parents, psychologists, and others meet to go over the results of the test and placement is recommended. If placement is accepted, an Individual Educational Plan (IEP) meeting is held to plan with parents, the special education teacher, and others the best objectives and methods for the child, and the child is placed. Access to appropriate education is guaranteed by law; a specific facility is not guaranteed.

Two important differences between access to compensatory education and access to special education should be highlighted. First, special education has an elaborate placement process that ostensibly ensures that the proper program is developed and that the parents of the potential special education student are involved in the decision making. For compensatory education, the placement system is much simpler and does not have the elaborate safeguards found in special education. Second, once a child has been defined as handicapped, access to special services is guaranteed by law. In contrast, a child from a poor family who is not doing well in school is not guaranteed access to compensatory education services. Access to compensatory education is limited by availability as well as eligibility.


A child that is mildly handicapped is entitled to appropriate educational treatment, almost always some form of special education. Such a child, on the one hand, carries the probability of rather permanent stigma associated with his condition, and on the other hand receives a more profound and deeply rooted sympathy for (and exemption from responsibility for) his condition. Further, a handicapped child rarely conveys any information about his family’s economic or social status.

A child who is a poor learner and whose parents are poor is eligible for compensatory educational treatment. Whether it is received depends on a number of elements, not the least of which is how many other poor children he goes to school with. A child receiving compensatory education usually receives it for only a few years at most, and tends not to receive a personal or permanent label such as “retarded” or “disturbed” that will follow him around. On the other hand, given that a student is eligible for compensatory education in part because of the economic status of his parents, he receives less sympathy for and exemption from responsibility for his or her condition. In summary, the recipient of special education gets with it a fairly permanent and sometimes very stigmatizing label; the recipient of compensatory education does not, at least to the same degree. A special education student is more likely to be seen as an unfortunate victim of fate deserving of all the help we can give, but incurable. A compensatory education student is more likely to be seen as an unfortunate innocent whose parents have failed in society and, partly, as parents, deserving of what we can spare. These separate conceptions tend to stand in the way of the educational problem that needs to be addressed, namely, how to provide the child with an appropriate and successful education.


The possibility of developing noncategorical remediation for the mildly handicapped has received some recent attention from practitioners and researchers.36 In at least three states, the idea of noncategorical remediation is state policy. The arguments for the noncategorical approach to special education mirror, to a certain extent, the arguments for the merger of compensatory education and special educational programs, which will be discussed below. The arguments for noncategorical special education center around the following issues: stigmatization of labels, accuracy of selection, irrelevancy of categories for instruction, and problems in providing for all categories in local schools.

The stigmatizing impact of special education labels has received considerable attention in the psychological and educational literature. It is believed by many that special education labels have potentially negative effects on students.37 They can affect the expectations of teachers and the student’s self-esteem. These effects are considered to be particularly strong if the student has been labeled EMR. While noncategorical programs for children in this group will not completely eliminate the possible stigma associated with receiving a remedial service, it is possible that the impact of the stigma may be reduced if the student does not carry one of the traditional labels.

Another negative aspect of the three-category approach is the inaccuracy of the identification process. This inaccuracy derives from a number of problems with the nature of the categories and the process used to place children in them. Problems in classifying students stem from

ambiguity in the definitions themselves (at state and federal levels)38

difficulty in measuring various disabilities precisely39

similarities in the symptoms exhibited by children referred40

variations in interpretations of specific data used for classification because of variations in psychological and educational theory held by persons doing evaluations41

As this list suggests, the basis for the classification of a child in a given category can vary considerably from state to state and from district to district. Such variation means that the probability of some misclassification (either false positive or false negative) is great indeed. These problems with identification have also resulted in large variations in estimates of prevalence and sudden changes in the number of children identified when the identification policies have been changed.42

The most significant argument for merging the categories is that they do not imply differing instructional approaches. This is due to two factors: One, there is the significant behavioral overlap among the three groups that has been noted: and two, the labels are of children, or at best their problems, not their treatment. A teacher cannot assume that an LD child has only a specific cognitive processing problem and not an emotional problem as well. Similarly, an EMR child may have emotional problems that must be dealt with and important patterns of strengths and weaknesses. More important, even where a label does give a teacher a better idea of the nature of the problem, the information rarely makes a difference in the instructional approach. Hallahan and Kauffman assert that the same types of materials are used for these three groups.43

One last point for noncategorical programs was recently raised by Reynolds.44 It relates to the requirement in P.L. 94-142 that a child should receive his education in the least restrictive environment possible. That is, the preferred location is the regular classroom. If that is not possible, some arrangement that is within the home school building should be tried. Only if it is absolutely necessary should a child be taken out of the home school. It is clear that not all schools have enough students to justify special programs in all the mildy handicapped categories. Therefore, children often have to be moved out of their home schools to centralized locations where they can receive instruction in either an LD, SED, or EMR program. On the other hand, most schools could support one noncategorical program. This arrangement would enable these children to remain at the local school.


The existence of special education and compensatory education delivery systems as separate structures has occurred for a number of reasons. We have discussed the differing historical evolution of each system that was the original basis of separation. We turn now to the three major rationales that support the separation of special education for the mildly handicapped from compensatory education: political, professional, and instructional.


Hobbs notes in the special education context the wide variety of social functions fulfilled by education labels: “to maintain the stability of the community and of its institutions, to control the allocation of resources and govern access to them, to reduce discord in school and neighborhood, to preserve majority values and expectations.“45 While all of these functions are in some respects political in nature, the one most clearly so is that concerned with the distribution of resources. Classification schemes contribute directly to the ability to mobilize support for and to allocate national and local resources to address the needs of students. These schemes also permit the public distribution of finite resources with minimum conflict.

Whether the label is “disadvantaged,” as in the case of compensatory education, or “handicapped,” “ mentally retarded,” or “learning disabled,” in the case of special education, labels provide an evocative basis for obtaining public recognition. In this sense the political function of labels has served a key role in targeting billions of dollars over the past fifteen years for the assistance of students perceived to be in some need.

The groups that pressed for compensatory education used such ideals as the elimination of poverty and the equalization of educational opportunity in their arguments for the funding of the programs. These groups were composed mostly of black activists and white liberals. The ideals that were used to support special education services included such notions as “the personal rights of the handicapped to a public education,” and the humanitarian responsibility of the community to help the “afflicted.” The organizations that lobbied for the programs were comprised mostly of the white middle-class parents of handicapped children. Without the involvement of all of these groups it is unlikely that as much money would have been appropriated to support the programs. Further, without the continued efforts of these groups, the continuation of the programs is in some doubt.

In light of these points, an important question is, Could these groups continue to effectively pressure legislators for support if the systems were combined? The ideals that would motivate a combined group would have to be somewhat different. Specifically, the ideals would have to focus on the rights of all children to receive the education they need to function well in society. However, it may well be that those supporting compensatory education may be reluctant to have their children served in a program that includes children with a personal disability. On the other hand, supporters of special education may be reluctant to have their children educated with minorities and the poor. It is possible that as these groups began to see themselves as having common interests, they could combine forces. But it must be admitted that this move toward a vision of common interests is problematic and unpredictable, especially in the immediate future, which seems to portend retrenchment in federal involvement in social programs.


Labels and classification schemes serve as markers of professional boundaries. Training (and in many instances compensation) in the profession, the issuance of degrees, and even the basic vocabularies reflect the various labels. Professional identification has been particularly important in the special education context. Many states require specific certification to teach in special classrooms. Some states even certify within particular special education classes. The presumption behind differential training rests on the notion that teaching special classes of students requires special preparation (e.g., specialized knowledge about the problems of these students and their treatments). This specialized background has tended to construct barriers between special education teachers and regular teachers, as well as between special education teachers and parents, that are somewhat analogous to those between doctors and lay people.46

Compensatory education has not fostered quite the same professional specialization that has been found in special education. The reasons are many. Perhaps the most important rests with the circumstances of the implementation of the first compensatory education programs. Both Head Start and Title I programs were implemented with very short lead times and at a considerable level of participation. In a single year hundreds of thousands of students were involved in these programs. This left little time to develop, let alone actually implement, specialized training or certification requirements. Great Society programs typically were put into place in an atmosphere of crisis and great expectation that worked against systematic, planned implementation, whereas special education programs have grown over a number of years, providing time and expertise with which to define professional standards.


The instructional rationale for separate labels follows from both the early use of the medical/clinical models in special education and political commitments of compensatory education-interest groups. Basically, the argument is the following: If the underlying causes of a problem are known and observed to be diverse, then the manifestations of the problem are either different or only superficially similar, and the proper amelioration or instruction for problems must be different.

Presumably, compensatory education children perform poorly in school because their life experiences have not provided them with the kinds of interactions that develop competency in school-related tasks. Specifically, they tend not to learn “how to play school.” The argument includes the following list of supposed deficits: (1) they tend not to have used the same abstract classification organizers (such as shape, color, size, mapping, measuring, comparing, etc.) in their daily lives as those used in school so learning school tasks is very hard; (2) they have folk knowledge that conflicts with technological mainstream knowledge so that social studies and science are not easily absorbed; (3) they also fail at arithmetic and reading, which leads to failure in the other subjects. Faced with this experiential deficit, compensatory education is designed to help students by providing them with both the missing experiences and the academic tools. In practice this translates into enrichment activities; field trips and neighborhood helper visits; and, most important, training in underlying cognitive skills or remediation in specific academic areas such as reading or mathematics.

Special education students perform poorly in school because their intellectual equipment is to some extent faulty. Their life experiences are assumed to be the same as (or comparable to) those of children who do well in school (SED children may be an exception here), but the student’s ability to cope with the usual amount of confusion in lesson organization and presentation is considered to be less than average. Faced with this equipment deficit, special education is designed to help students by setting clearly obtainable objectives, mastering those, and going on to the next set. In practice, this translates into instruction in perceptual, cognitive, and social skills, and remediation in reading and mathematics.

The instructional rationale generally assumes that it is better to treat root causes rather than symptoms. Remission is never as good as a cure. The special education profession has placed strong emphasis on diagnosing the causes of performance difficulty in order to appropriately adjust treatments or goals. Similarly, compensatory education seeks to cure the root cause of poverty in successive generations by breaking the cycle that witnesses each new generation of poor growing up without the prerequisite skills to obtain meaningful employment. Here the root cause of poverty is not the absence of employment but the absence of employability.

This rationale is sensible in instructional terms only if significantly different instructional treatments and expectations are required for the treatment of various root causes. However, if it is possible that there are different causes that present mildly different manifestations but that these respond to essentially similar treatments, then etiology alone cannot be used to justify separate classifications for instructional purposes.

A comparison of the treatments shows that they are very similar for both groups. This fact weakens the position of the argument for separation. It is possible, however, that a closer inspection of the effective elements of the treatments would reveal that different types of instruction work best for the two groups of students. That is, details of effective instruction might still be different in ways that overlap group membership. However, if the best treatments for specific problems are the same or are different on bases other than labeled group membership, then the existence of differing etiologies would be irrelevant. From an educational point of view, the instructional rationale is the only possible reason for separate systems.

Interestingly, the instructional rationale, although implied in special and compensatory education structure, is almost never directly addressed in the literature. The central argument of this article is that there is no justifiable instructional rationale for separation. On the contrary, we have found that there is a remarkable overlap in the successful educational practices of both groups. Hence, there is no educational need to label children as compensatory or special education students and to serve them in separate systems.

We take the position that, in spite of slightly differing instructional goals, the educational practices for both groups tend to be similar. This means in effect that the instructional rationale neither provides a logical basis for program separation nor in practice does it provide a functional one. Evidence for this position is presented in the next section, in which we examine the major areas of instruction for both compensatory education and special education students and review the instructional methods that have been found to be successful for each area.


The assertion that the effective practices for compensatory education and special education are essentially the same rests on a review of effective practices for both groups. For special education, we reviewed all the published evaluations of programs for the mildly handicapped that we could find. We examined the subset of programs that gave solid evidence of success. We took as evidence of success gains in standardized reading and math achievement tests and/or measures of social-emotional growth. With respect to academic achievement, we took a gain of one grade-equivalent-year in one academic year of instruction as a minimal criterion for program success. We then listed the major features of the programs. These features were the elements that either the designers of the programs designated as important or we thought were particularly characteristic of the programs that were described.


Mildly handicapped special education students receive instruction in two different settings, self-contained classrooms and resource rooms. In self-contained classrooms there are nine to fifteen students with one teacher and maybe an aide, depending on the disability. About 60 to 70 percent of the day is allocated to reading and language arts with the remainder divided between math, social studies, life skills, science, and some nonacademic subjects. In resource-room programs, students are in regular instructional settings except when they go to the resource room for reading or math instruction. In the resource room there are usually about six students. Both settings tend to have strong behaviorist approaches to management and instruction. These approaches include overt reinforcement systems, small modular instructional sections for which mastery assessment is made.

Compensatory education students receive 85 percent of their reading instruction and 54 percent of their math instruction in a “pull-out” situation similar to resource rooms.47 However, when the students are pulled out they may join other students in a setting containing more than twenty students. Other compensatory education students are in classes in which all the students are compensatory education or in which an itinerant teacher visits and tutors. The main content of extra instruction is reading and math with minimum time spent in counseling or in life-skills instruction. The most successful compensatory education programs (in terms of academic growth in basic skills areas) have tended to have behaviorist elements, but they are by no means the majority.48

In the next section we review instructional practices that alone or in combination have been shown to be especially helpful in increasing the academic performance of children who need help. Some of these variables (e.g., small class size) are important because they increase the likelihood that other important activities will take place (e.g., increased time receiving cognitive instruction). Other variables are important in and of themselves as instructional treatments. Some variables, mentioned primarily with one body of students and not others, we presume to be important for both.

Small Class Size

Class size is a variable with a tangled and interesting past. For years, researchers have claimed that smaller classes might make teachers feel better but that they did not in any way affect achievement. Special educators have tended to ignore this finding. Self-contained special education classes are almost always smaller than average, ranging from nine to fifteen children depending on state, age, and label. Resource rooms almost always have fewer than ten students for any instructional period. Thus, special educators assume the importance of small classes.49

The rest of the educational research community became convinced by the publication of the meta-analyses of Glass and Smith.50 Their extensive review from 1900 to the present notes substantial advantages as class size drops from forty to twenty to below twenty. These advantages are in terms of both academic achievement and affect for regular and compensatory education students. Small class size does not interfere with any other valued instructional technique. Clearly, children that need help in school should receive that extra help in as small a group as possible. However, it should be noted that both successful and unsuccessful programs in special education had small class sizes; thus, small classes do not guarantee success.

High Content Overlap among Teaching, Learning, and Criterion

Content overlap is perhaps the main element of direct instruction.51 When teachers spend their time and energy teaching students the content that the students need to learn, students learn the material. When students spend their time actively engaged in activities that relate strongly to the material they will be tested on, they learn more of that material. Evidence supporting this comes from the Follow Through evaluation,52 from Beginning Teacher Evaluation Study (BTES),53 from Instructional Dimensions Study (IDS),54 from a study of reading in learning disabilities classes,55 and from several other smaller studies. This is true for both compensatory and special education.

The difficulty with direct instruction is, “direct instruction in what?” If students need to learn to read orally or silently, they must spend time reading rather than discussing reading (language experience); however, if students are to be tested in discussion or essay writing, they must spend time doing those things. Direct instruction in criterion-relevant material is in competition with more humanistic and indirect approaches to learning. Unlike class size, direct instruction does interfere with the simultaneous use of other valued approaches.56


Both direct and indirect evidence exists for the efficacy of mastery learning. The notion of formal assessment on completion of blocks of instruction is prevalent in many of the more successful instructional programs used in compensatory education.57 Most of the special education programs use some form of mastery learning.58 Further, considerable literature exists that suggests that mastery learning is a successful approach to instruction.59 Mastery learning has a cost in that, like direct instructional approaches, it tends to narrow the range of what is taught to those things that are easily tested. Interestingly, while evidence is accumulating about the positive effects of mastery approaches, no similar evidence is accumulating on the significance of diagnostic assessment, although many programs contain diagnostic components.


A third component in any instructional intervention is the amount of time devoted to the particular area of interest. The issue of time is confounded with many themes—time allocated to an area of work, time allocated to special instruction, time spent attending to task, time spent attending to criterion relevant tasks, cost of time away from regular instructional settings. For both compensatory education and special education, an increase in time on criterion-relevant material has direct and powerful effects. Further, any device that increases the time spent receiving direct cognitive instruction seems to have strong payoff in terms of increasing the amount of time a student is actively engaged in learning and therefore the amount of material learned.60 This increase of time spent receiving instruction from a qualified teacher can be most easily accomplished by reduction of class size and/or skillful and fluid grouping practices. What is important to emphasize is that time spent with a teacher on material that is not criterion-relevant does not have payoff for that criterion. The same is true for time spent doing tasks that have low relevance for the criterion.


An area of instruction that is entwined with time is pacing. The natural tension that exists within pacing is to allow enough time to get the material learned (lowered pace) while moving rapidly enough not to get bogged down (increased pace). Students do not seem to be the best judge of this balance.61 In general, the more rapid the pace, the better off the student is,62 but pacing is not powerful enough to be important in the presence of other explanatory variables. That is, the importance of pacing, at least in reading instruction, “washes out” in the presence of overlap and time spent in direct reading. However, pacing is an easily manipulated instructional variable and one that is probably important for teachers to concern themselves with.


Both special education and compensatory education programs emphasize the importance of motivating the student to want to learn and to engage in behaviors likely to increase learning. Evidence for the specific practices that are useful in increasing motivation is sketchy and contradictory, but evidence for the functions that need to be filled is considerably stronger. Basically, any system that delivers reinforcement as a consequence of students’ performing learning tasks well tends to increase the likelihood of those behaviors increasing. There are multiple configurations with respect to the type of reinforcer, the quantity of reinforcement, the schedule, and so forth, but both the compensatory environment63 and the special education environment64 show positive effects for positive reinforcement procedures. Other techniques for increasing motivation are often discussed (capitalizing on high-interest materials, for example), but evidence for their use and/or success is unavailable. Motivational systems tend to be more elaborate and more tightly contingent in small classrooms, either self-contained or resource rooms. Installing a moderate program of reinforcement does not seem to interfere with any other valued goals.

Links between Remedial and Regular Instruction

Special educators are generally in the position of initiating and establishing links as the students in special education programs are mainstreamed. In contrast, compensatory education teachers are usually on the receiving end of shifts of pull-out students. Regardless of which position the teacher is in, both categories consider solid substantive and scheduling links between remedial and regular instruction to be important.65 The disturbing descriptions of some students’ days in which they are on a roller coaster of “special” assignments is only one consequence of failing to provide for these links.

Contact with “Normal” or Regular Pupils

Both special education and compensatory education have stressed the importance of integrating “special” children with “normal” children.66 This emphasis has occasional support in the research literature67 and in the literature on mainstreaming.68 There are two important points. First, it is socially and morally desirable for children to be educated in integrated environments. This desirability constitutes an end in itself and does not require justification in terms of academic improvement. Second, it is also desirable for each child to be given the type of education necessary for his maximum growth and that type of education may entail some degree of segregation. But that segregation should be absolutely minimized. Pursuing a policy of contact with normal students may interfere slightly with other valued educational practices.


From the review of successful remedial practices, a picture of an effective remedial classroom has emerged. Students are taught either one-on-one or in very small groups (three or four). The material is broken up into small pieces, with the ‘teacher carefully introducing each new piece and modeling the correct responses. Monitored immediate practice follows with attempts to limit or eliminate student errors. Extended practice of the new skill in gradually more complex settings follows the monitored practice. Prompt feedback on performance follows periodic review, and integration of new material follows successful grasping of the small “new bit” of information. A final element in the description of successful programs is the linkage between the remedial help and the regular teacher.69


Up to this point, we have detailed the development and forces behind both compensatory education and special education. We have reviewed in some detail the broad structures of supplementary education that seem to be most successful for cognitive growth in basic skills. In spite of the similarity of treatment and in spite of only noninstructional reasons for separation, there is no mandatory need for merger. What does seem to make a compelling case for such a merger is the combination of these reasons with the rather lengthy list of negative effects of labeling: misclassification, poor treatments, and poor dollar-spent-to-dollar-delivered ratios.

First, there are presumed and actual problems associated with the labeling of children. By far the worst label is Educable Mentally Retarded. It is this label and to a great extent the expectation (or lack of it) for children so labeled that has caused the greatest concern.70 But all children that are labeled are to some extent stigmatized.71 Labels often limit the level and depth of treatment and can create a barrier between peers.

Second, as long as children rather than the treatment receive the label, and as long as there are ranks (real or imaginary) among the labels (EMR, SED, LD, Compensatory Education), serious problems arise for children who are misclassified. Considerable evidence exists that children with the same behavioral profile living in different states would receive drastically different classifications.72 Further, the reliability of psychological placements across psychologists is low,73 so that a child may receive a different label depending on the assessor. The consequence of such mislabeling has been considered serious enough for the courts to rule on the appropriateness of the mechanisms of classification.74

Third, there is the very real problem of the disproportionate presence of black and bilingual children in some categories, especially EMR, and the concomitant disproportionate absence of these same groups in other classes such as LD.75 The problem of misclassification is one associated primarily with special education, but to the extent that there is some overlap, there is the potential for misclassification from compensatory education to special education, although not the other way around.76

A fourth problem is that after a decade and a half of experimentation with different types of treatment, something is known about effective modes of instruction, and these are not reaching as many children as they could, especially within compensatory education. Specifically, compensatory education spends the most amount of its treatment dollars on paraprofessionals. The presence of a paraprofessional in a classroom of thirty children reduces the student-to-adult ratio from 30: 1 to 15: 1; it does not reduce class size by half. Nor does it guarantee that students are receiving the kind of quality treatment they might if they were being taught by a trained teacher.77 In addition, the focused, carefully constructed curriculum that is designed to improve reading, mathematics, and writing tends to be diffused in some classrooms into less intensive, less-focused efforts in areas in which children do not need practice. A merger of programs might permit a more effective concentration of instructional practices for all students.

The fifth problem associated with the current separation of programs involves the quantity of money spent in relationship to the quantity of money received. While no figures are available for Special Education, the figures for compensatory education are disheartening. For the $1,000 spent per compensatory education child, only about $400 actually reaches him. Much of the remaining $600 goes to supporting a complex system of service delivery that is vainly trying to make certain that the wrong child (i.e., noneligible) does not get the money.78 In compensatory education, money follows the institution, not the child, so that in a population with known high mobility a child may bounce in and out of remedial services frequently in the course of his educational experiences.79 It would seem far simpler to recognize that approximately 20 percent of our nation’s children have some real educational difficulties and make special funds available to that group more directly than is now done. In such a merged system, duplication of administrative costs and burdensome bureaucratic oversight costs could be reduced.



The purpose of this article has been to examine compensatory and special education programs with the following question in mind: Can the legitimate educational needs of low-achieving students be more adequately and efficiently met by a single system (without categorical labels for the child) that serves both groups of children? The separate origins of compensatory and special education programs have been traced. In addition, rationales (e.g., political, professional, and instructional) that provide a basis for maintaining their continued separation as discrete programs have been reviewed. The rationale most relevant to the question of serving “legitimate educational needs,” the instructional rationale, has been examined in light of recent research on what constitutes effective instructional practice, regardless of the type of student or the program context.

A review of the research on effective instructional practices (where effective is taken to mean having powerful impact on basic skill performance) indicates that practices that work for one classification of slow-to-learn students also work for other classifications of these students. That is, process variables and treatments such as smaller classes and direct instruction are effective across regular, special, and compensatory education classrooms.

The conclusion that effective instructional practices are “category-and program-free” has important implications for the instructional rationale that maintains separate systems. It is our position that the maintenance of separate systems is not defensible on instructional grounds. This position is further reinforced by two additional facts. One is that studies have found marked similarities among the populations assigned to various categories.80 A second important fact is that the current systems for labeling, especially those that label the child rather than the treatment, can have negative effects on the children served. Similarity in effective treatments and in populations served and the potentially negative consequences of labeling are persuasive arguments against the maintenance of separate systems for treating students currently not achieving in regular classroom settings.

The suggestion to merge specific special and compensatory education programs on educational grounds is not to belittle the power of political and professional rationales for their separate maintenance. A merger will be difficult to achieve under the most favorable of circumstances. It is open to question whether noncategorical aid can provide the basis for a coalition of public support to the same degree that the current emotion-laden categories have done in the past. Professional resistance to mergers will also most certainly develop. One answer to such opposition may come from a change in the training of the various professional groups. In the context of a newly envisioned program designed to meet specific educational deficits, individuals will be trained based on the treatment they are expected to give (e.g., reading or math remediation). Under these circumstances, one would not have a special education or a Title I teacher, but specialists in teaching mathematics or reading.


In taking a position that the systems be merged, we also admit that such a merger involves certain risks. A central one is whether a new system will be as politically viable as the two separate ones in marshaling resources for the children involved. We simply cannot definitively answer this question. The vitality of a new system would rest on the various interests’ recognition of common needs. (It may also be true that a new system would require less resources, reaping the benefits of a reduction in administrative costs required to maintain dual distribution bureaucracies.)

A second concern must be with the specialized training that currently is maintained through the present structure. Frankly, we see a ready response to this issue. A new system would presumably be problem- not label-oriented and specialization would presumably continue under such an orientation only more effectively (directly) linked to instructional issues. Such a functional approach to specialization in training could be suitable across the gambit of problems likely to be encountered and not simply to cognitive growth in basic skills.

A third concern rests with the multiple goals that the systems have had beyond classroom instruction. One example might be the parent involvement (parental advisory committees) that has played an important role in the evolution and impact of Title I programs. This type of involvement represents a major innovation for many districts serving poor and minority communities, providing unprecedented access to the decision-making process. It would be urgent in any merger to take care that such significant innovations are integrated into the new system. The instance of parent involvement may present little difficulty in that both systems contain strong legal provision for this, and a merger would constitute a bit of working out the details rather than the reconciliation of incompatible philosophies.

Fourth, we have cited evidence in the area of cognitive growth in basic skills to challenge the instructional rationales of the currently separate systems. More evidence is needed in this area and additional evidence must be marshaled in other instructional areas (e.g., social growth, behavioral modification, etc.) to assure that our position that what tends to work in one program area will also work in other programs is correct.

Despite the examples of concerns we have listed above, we remain convinced that the current rationales for separate labeling and treatment systems have serious flaws. At the very least, we feel that the instructional evidence is such that the burden of proof should be shifted as to why the separate systems make sense.

Any suggestion for a major shift in current educational policy brings with it the responsibility of describing what the future policy should be. A detailed description of a merged program is beyond the scope of this article. However, the essence of what is envisioned has already been described. A merged program would label the treatment(s), not the child. A label such as Extraordinary Educational Service (EES) might do. Obviously, this does not remove totally the negative effect of being selected for treatment, but it does dilute the sense of permanence historically associated with something such as EMR in both the public attitude and the professional treatment. Furthermore, the problem of locating a deficit in the individual is somewhat ameliorated. Any program name cannot avoid some labeling effect. It is only hoped that with a designation such as EES, the negative effects for individuals can be minimized.

The philosophy and goals of a program of EES should combine the best in the philosophy and the goals (and program characteristics) of both special and compensatory education. Namely, each child should be given sufficient instructional support so that he can comfortably benefit from regular instruction. The remedial instruction should be aimed at helping, a child catch up and cope with regular instruction.

On the one hand, the ideal EES would be readily and easily available to any child that needed it. On the other hand, past experience suggests that any program of this type runs the risk of becoming a dead-end dumping ground for children not wanted by the mainstream system. Therefore, we would recommend that the current twenty-day emergency program placement be used in combination with the cumbersome but safeguarding system of due process. What must be guaranteed is that parents are informed formally that their child is recommended for the receipt of EES and that they continue to have a meaningful role in the decision-making process. Receiving such services should have a time limit and a performance objective (e.g., one semester and reading at the same level as the bottom reading group). Performance must be measured in terms of the criteria for placement and it must be linked to the ability to exit the treatment program.

A student leaves a compensatory education program for one of the following reasons: The child moves and changes school; the child changes grade; the cutoff is rearranged within a school district or school; the child progresses beyond the cutoff, or the child is promoted to an age/group with no services. Special education, with the current pressures of P.L. 94-142, engages in long good-byes. A student may improve dramatically and be recommended for partial mainstreaming in the area of improvement if the mainstream class has room for him and if it has minimal existing administrative cost. A student may be partially mainstreamed because of parental pressure. A student may be mainstreamed because he graduates into an age/grade level without services or with more restrictive services. But neither special nor compensatory education have planned exits contingent on student improvement. There are financial and administrative pressures against exiting. It is recommended that both access and exit be matched to student performance in a clearly identified area, either academic or behavioral.

Finally, EES would be linked closely to the regular classroom. Responsibility for the programmatic development for a child would never leave the classroom teacher. Ultimately, an improvement in regular classroom techniques may diminish the necessity for EES for many students in the first instance. This aside, it is essential that a student receiving EES and returning to the classroom not be left to flounder in ways that will undo the effects of such a program. It is hoped that the provision of EES to students in need can provide the delivery of effective instruction to those most in need, with potentially lower costs and reduced opportunities for negative labeling.

We do not pretend that the proposal for merger, whatever the specific characteristics of the new system might be, would be easily accomplished in the current political and educational context. However, we remain convinced that the rationale for separate systems is seriously flawed, and that the goals of both systems may be better accomplished through a new strategy that concentrates on the characterization of the service, and not the individual, as its essential strategy for the delivery of assistance.

Cite This Article as: Teachers College Record Volume 84 Number 2, 1982, p. 391-422
https://www.tcrecord.org ID Number: 834, Date Accessed: 1/23/2022 5:26:41 PM

Purchase Reprint Rights for this article or review
Member Center
In Print
This Month's Issue