Printer Friendly

Categorical programs for remedial and handicapped students: issues of validity.

Categorical Programs for Remedial and Handicapped Students: Issues of Validity

In this article, we examine the instructional validity of current educational policy regarding categorical programs serving low-achieving students. We raise questions about the relative instructional level and learning rates of learning disabled (LD) and other remedial students, and about the instructional methodologies best suited for these student groups. These questions lie at the heart of the rationale for categorically organized special programs for low-achieving students and the means by which such programs can optimize their services, concerns that are central to the Regular Education Initiative (Will, 1986). Examination of the instructionally relevant learner characteristics of LD and remedial students will help address issues such as the extent to which LD and remedial programs serve children with different instructional needs; the feasibility of grouping LD and remedial children together for instruction; and whether the majority of low-achieving, special needs children could be served within one integrated program with one funding source, one set of rules and regulations, and one administration rather than in a host of distinct categorial programs as they are now served.

Before describing the problems with the present system, we want to emphasize that these problems do not detraact from the worth of individual programs nor from the validity op the reasons that led to their establishment. The fact that solutions to one set of problems create sill other problems may be inevitable in the evolution of any system. Yet, it is no accident that task forces, federal initiatives, and discussions in the literature now focus on the lack of coordination among services to low-achieving children and ways of addressing that problem. These discussions are also occurring among administrators and teachers at the district and building levels. There is a healthy questioning of the validity and usefulness of the current relationship among separate, narrowly defined programs and a willingness to entertain the possibility of cooperation and integration of programs and services. A paramount issue is whether the present systm provides the best possible instruction for children and enables teachers and administrators to provide their services with optimal effectiveness and efficiency.

Categorial Programs

Instructional problems posed by individual differences are a fact of life in classrooms. By some estimates, as high as 25% of elementary-aged students are considered unable to learn basic skills under ordinary classroom conditions (Birman, 1979). As a means of easing the instructional problems that arise from student diversity, lawmakers have created a host of special programs. These categorical programs--special education, Chapter I, and state and local district remedial programs--were created to meet specific needs: providing an appropriate education to children with disabilities, overcoming the effects of poverty, or assisting low-achieving students (Leinhardt, Bickel, & Pallay, 1982).

As they now exist, each program functions separately, conforming to regulations that discourage cross-program integration and the sharing of instructional resources. Students with learning disabilities are taught in special education programs by special education teachers, students with limited English proficiency are taught by teachers funded through bilingual programs, and remedial students are taught in various remedial programs sponsored by the federal government (Chapter I-Disadvantaged and Migrant), by state government, or by local school districts. The introduction of categorical programs, each created at a different time and emphasizing the needs of a specific target group, followed a pattern that Reynolds and Wang (1983) characterized as disjointed incrementalism, rather than pursuit of a comprehensive program that addressed the needs of all or at least the majority of students experiencing difficulty learning in the classroom.

Services for low-achieving children in elementary school are characterized by an array of eligibility requirements, rules and regulations, and accounting systems. The result is often a sense of fragmentation at many levels. Programs which may be quite similar at the teacher-child level are managed by different administrators, implemented by different teachers, and provided only to certain chidren who, in fact, may be more similar to than different from children in other programs. If special teachers are funded by more than one program, then strict accounting of time and services is required to satisfy regulatory demands. When these programs use a pull-out model, the classroom resembles a rail station, with arrivals, departures, groupings, and regroupings--the teacher acting as a dispatcher and students traveling to different destinations. A child might receive remedial help in spelling from one program, reading in another, math in still another. While there is nothing inherently wrong in providing diversity of instruction to children (and in some cases this may work to the child's advantage), the "system" as a whole lacks rational coherence.

Fragmentation of Services

Fragmentation of services takes two forms: (a) between general education and special programs, and (b) among special programs. Both have been discussed in the special education literature, most often with emphasis on the former (Reynolds, Wang, & Walberg, 1987; Stainback & Stainback, 1984). We must emphasize that the fragmentation referred to in this article is within special programs, the group of categorical programs designed to serve specific populations of children with mild to moderate learning or behavior problems, whatever the etiology or genesis of those problems (e.g., cultural differences, economic disadvantage, learning disability). The issue under discussion here is not the merging of regular and special programs into a single system or subsuming one to the other. Neither do we focus on mainstreaming per se or the specific environment where special programs are delivered (e.g., the regular classroom, resource room, or remedial reading room), or the different service providers (classroom teachers, special or remedial education teachers, aides). Rather, our focus is on integration of special programs providing instruction supplementary or complementary to, or supplanting, general education provided by classroom teachers. The fragmentation of these services is the presenting problem, and it is difficult to defend.

In our view, the success of any effort to coordinate and integrate services for low-achieving children will depend on addressing and resolving two general validity issues. The first in instructional, the second is political. In this article, we will concentrate on the first issue, with data from recent studies suggesting that there is instructionally valid reasons for integrating services to low-achieving children. We then discuss the political validity of establishing a unified program.

INSTRUCTIONAL VALIDITY

What is instructionally more valid: the present system of discrete special programs or a reformed system of services where all low-achieving students can be taught under one coordinated, unified program? Put another way, do the students taught in the various categorical programs have different enough instructional needs to justify separate programs, or are they similar enough to warrant noncategorical grouping for instruction?

Many child characteristics can affect teachers' decisions about instructional grouping. The pedagogical literature is top-heavy with rhetoric about the importance of recognizing individual differences and the necessity to tailor instruction accordingly. The number of measurable human traits approaches infinity, and measurement of any of these traits will result in a range of scores. Teachers who attempted to individualize instruction according to each student's traits and preferences would be paralyzed by the requirement for comprehensive pre-instructional assessments, not to mention the planning required to use those data in designing instruction.

Although the list of potentially measurable human traits is overwhelming, in practice we need ask but one question: On what traits do individual differences create different instructional needs? This question reduces a myriad of traits to three instructionally relevant trait categories: instructional level, learning rate, and learning style.

The instructional validity of the "unified program" concept, that is, integration of remedial and special education, depends on answers to three questions: (a) Are students from the various groups at different instructional levels? (b) Do students from various categorical programs learn at different rates to such an extent that their progress would be hindered if they were taught together? and (c) Is the type of instruction appropriate for one group not appropriate for the other? A discussion of these issues follows.

Instructional Levels of LD and

Remedial Populations

According to Slavin (1987a), possibly the most difficult problem of school and classroom organization is providing instruction at an appropriate level. He defines appropriate instructional level as the situation where "the teacher makes sure that students are ready to learn a new lesson (e.g., they have the necessary skills and knowledge to lear it) but have not already learned the lesson" (p. 92).

The organizational problem referred to by Slavin is how to strike a balance between teaching each child at the appropriate level and yet provide instrucion to several students at the same time. Grouping elementary-aged students according to instructional level is an educational practice born of conventional wisdom and supported by research (Slavin, 1987b). Do categorical special programs maximize appropriate groupings of students for instruction--that is, do their divisions correspond with the divisions that would be made by a teacher attempting to separate students according to instructional level; or do categorical programs artificially separate students whose instructional level is similar?

No previous research has directly addressed the comparability of LD and other low-achieving students' instructional levels. The closest relevant research we could locate were studies that compared the mean achievement of various student types. This research has produced a somewhat confusing picture partly because different studies have used different types of measures (e.g., commercial achievement tests vs. cirriculum-based assessments); compared different research populations (e.g., regular education, special education, Chapter I, low-achieving, and LD students); and employed very different sample sizes (from as few as 15 to more than 600 students per group). Two studies reported large mean differences between groups (Shinn, Tindal, Spira, & Marston, 1987; Shinn, Ysseldyke, Deno, & Tindal, 1986); another study indicated statistical but little practical differences between the groups (Ysseldyke, Algozzine, Shinn, & McGue, 1982); still another study discovered significant differences at some but not all grade levels (Marston & Magnusson, 1985; and one study revealed no significant differences between group means (Candler, Johnson, & Green, 1983).

We will describe two recent studies that examine, more directly we believe, the comparability of instructional level of students served in special education and various remedial programs. In the first study (Jenkins, Peterson, & Schrag, 1988), we examined more extensive samples of these populations than were described in earlier studies, using not only a larger sample, but also one drawn across grade levels and school districts. Moreover, rather than using an arbitrary achievement score cutoff to define low-achieving nonhandicapped students, we functionally defined this group as all students whom schools served in remedial programs. Similarly, LD students were defined as those students identified as LD and served by schools in special education programs. Most importantly, however, we used score distributions rather than mean differences to compare the two groups. In this way we hoped to determine the relative instructional levels of children served in adjacent but separate programs within the same schools.

In the second study, we examined the instructional groupings in one school district were students from various categorical programs were assigned to groups on the basis of instructional needs rather than by categorical labels. We wanted to know whether a more naturalistic than regulatory system for grouping--simply, at what level the children are performing in school rather than what eligibility criteria they meet--would result in categorically intact or mixed instructional groups. The extent to which children from different categorical programs are successfully placed together for instruction can be taken as one indicator of their similar instructional needs.

Study I

This study used performance on standardized achievement tests as a proxy for instructional level. To precisely determine instructional level usually demands measurements that are curriculum referenced. However, because performance on achievement and curriculum-referenced tests is highly correlated, with coefficients in the range of .80 to .95 (Deno, Mirkin, & Chiang, 1982; Fuchs & Deno, 1981), instructional level can be roughly estimated by examining students' achievement levels.

Subjects. The original sample included over 7,000 students from grades 1-6 in 28 schools (six school districts). The tested sample included 276 LD students, 955 remedial students, and 6,496 regular students. LD students were included if they were served in regular classrooms for more than 50% of the school day. Remedial students were served in either Chapter I, Washington's state-funded Learning Assistance Program, or local district remedial programs, which four of the six participating school districts operated. In general, students were considered "regular" if they were not served in a remedial or a special education program. Students were considered LD or remedial in a subject area only if they received instruction in that area. For example, students were considered LD in reading, but classified as regular in spelling, if they received special education services in reading, but not in spelling.

Procedure. In autumn 1985, all participating schools administered standardized achievenent tests to all students in grades 1-6. Students who were estimated to be performing at 2 years below grade level, or whose achievement was too low to be validly tested on the grade level tests, were given a test (off-level) below thier current grade placement.

Results are reported in Normal Curve Equivalences (NCEs), a statistic similar to percentiles; however, unlike percentiles, NCEs have equal interval properties. NCEs range from 1 to 99 and are computed in normalized standard deviation units. Test data for students given off-level tests were translated back to grade level norms so that the scores for all students within a grade level were comparable.

Results: Differences Between LD and Remedial Students. The mean reading NCE of the LD group (22.7) was significantly lower than that of the remedial group (31.6) as was the arithmetic NCE (23.6 vs. 35.S), but spelling NCEs (18.4 vs. 20.8) were comparable. For details see Jenkins, Peterson, and Schrag. 1988.

Figure 1 shows the relative distribution of regular, remedial, and LD students' reading achievement and their overlap. To facilitate comparisons between the groups, we plotted percentages rather than numbers of students, because the number of students in the three groups greatly differed.

Despite the large mean differences, the amount of overlap in the LD and remedial distributions is striking, with the greatest overlap between NCE scores 15 and 50. In fact, the distribution overlap at every point along the NCE scale, suggesting that for every student in one group there are students in the other group who obtained a like score. Moreover, the overlap between the remedial and LD students is far greater than the overlap between regular and LD students or between regular and remedial students. Using achievement level, it is far easier to distinguish between LD and regular students than between LD and remedial students.

The answer to our question, "How similar are LD and remedial students' achievement--that is, instructional level?" depends upon how one looks at the two groups. Focusing on the means of the groups emphasizes their differences. Focusing on the distribution of the two groups emphasizes their similarities. Although the averag LD student scored lower than the average remedial student in reading and arithmetic, many LD and remeidal students had identical achievement levels. The results of Study 1 suggest that teachers could combine LD and low-achieving students to form instructional groups if they were to use instructional level as a grouping criterion rather than the students' category labels.

Study 2

A second way to investigate correspondence in instructional level among students served in various categorical programs is to examine actual instructional groups of students who are served in an integrated program. School districts are strongly inclined to keep programs separated rather than integrate them because the maze of specific program regulations, standards, and fiscal tracking requirements poses a significant diterrent to integration. If they bypass regulations for the separate programs (fiscal accounting, teacher certification, and eligibility criteria), they do so at their peril. Mukilteo School District in Washington State is one that has succeeded in coordinating categorical programs at the level of instructional services in seven elementary schools (Felix, Hertlein, McKenna, & Rayborn, 1987). In this district, multiple categorical programs operate under a single service program; administrators use a detailed tracking system to maintain the legal requirements of individual programs. Students from various categories are grouped for instruction on the basis of curriculum-based measurements of instructional level, rather than category. Therefore, it is possible to determine the extent to which students from various categories in this school district program function at similar instructional levels by examining the composition of instructional groups.

Procedure. We examined the composition of every instructional group that was formed for students funded under the following special programs: special education, Chapter I, bilingual, the state-funded remedial program, and the district-funded remedial services. Reading, math, spelling, language, communication, and study skills were foci of instruction. The coordinated service model made extensive use of the District Instruction Programs published by Science Research Associates, for example, Reading Mastery and Corrective Reading. Teachers assessed students' instructional levels and assigned them to groups using the placement tests accompanying these programs. We selected for examination the instructional groups as constituted in December 1986.

Results. Table 1 shows the total number of discrete instructional groups across the seven schools. Out of 173 groups, 67 (39%) included at least one special education student. However, only 4 of the 67 groups (6%) were composed exclusively of special education students. Of the four groups that included only special education students, two provided instruction in study skills (these were the only study skills instructional groups in the school), and two were reading groups. Thus, when teachers used curriculum-based assessments to determine students' placement in a curriculum sequence, they formed homogeneous instructional groups that mixed students who had different categorical labels.

Taken together, these studies suggest that special categorical programs do not divide students according to their instructional level. Indeed, they appear to separate students who on the basis of instrucitonal level would be placed together. At the very least, it seems that a unified program for low-achieving students would not violate the element of instructional validity that calls for student groupings consistent with instructional level.

Learning Rate

We now consider the second instructional validity issue, learning rate. Do categorical programs divide students according to their learning rates so that students who progress at similar rates are grouped together? Alternatively, would a unified program inadvertently increase the chances that students who vary greatly in learning rate are detrimentally grouped together? This validity issue boils down to the question of the comparability of learning rates of LD and remedial students.

Measurement of this second trait category relevant to instruction, the speed at which students learn, requires repeated testing and is complicated by the fact that learning speed varies not only among students but also within individual students--a student might make faster progress in mathematics than in reading.

We were able to locate four studies that compared the learning rates of special education and remedial groups. None of the studies were designed expressly to compare learning rates of the two populations. As a result, the statistics reported in several investigations, mean differences, do not adequately portray the similarities and differences in learning rates of LD and remedial groups. Nevertheless, the findings suggest that, on the average, LD students progress more slowly than their remedial peers.

Marston and Magnusson (1985), using cirriculum-referenced measures, reported that Chapter I students' average learning slopes in reading were significantly steeper than those of their special education peers. Over a 16-week period, Chapter I students on the average increased their correct word reading by 2.8 words per week whereas special education students increased by 1.7 words per week.

Using similar measures, Shinn et al. (1986) compared LD and low-achieving students' acquisition rates for reading and spelling. Remedial and LD students increased their correctly read words by 2.35 versus 1.74 words per week, respectively, and increased their correctly spelled words by .64 versus .52 words per week, respectively. In contrast to the Marston and Magnusson (1985) results, the Shinn et al. (1986) differences were not significant, but this may have been because of the shorter duration of the latter study and smaller sample size.

O'Shea and Vacante (1986) studied LD students and low achievers longitudinally over several years using standardized achievement tests. Low achievers made significantly greater gains than those of their LD peers in math and language between grades 2 and 5. Gains in reading were not significantly different for the two groups, but favored remedial students.

Using the same population described in Study 1 of this article, Jenkins et al. (1988) compared the achievement gains of LD, remedial, and regular students over one school year. Remedial students gained over twice as many NCE points as LD students in reading (7.4 vs. 3.5, respectively), showed more than three times the growth in arithmetic (5.89 vs. 1.87), and more than twice the growth in spelling (6.07 vs. 2.49).

However, we believe that using group means to compare the average relative achievement gain of LD and remedial students is as misleading as using group means to compare the relative instructional levels of these groups. Mean differences mask the similarities between the groups. Figure 2 shows the overlap in annual achievement gains by students in the Jenkins et al. (1988) study. Although the remedial group on the average made larger gains than did the LD group, the similarities between the groups are more remarkable than the differences. The distributions of remedial and LD scores reveal large variability and considerable overlap, similar to that seen on measures of instructional level (e.g., in Studies 1 and 2), suggesting that many students in one group would have learning rates virtually identical to those of students in the other group.

In summary, is a mean difference in learning rate between remedial and LD students a serious obstacle to the creation of a unified service system for low achievers? No more so than is the mean difference between remedial and LD students on instructional level. At most, group differences in learning rates between remedial and LD students would suggest that teachers would on the average advance remedial students into higher groups more often than LD students. Teachers who adhere to a policy of flexible and fluid grouping can periodically reassign students to instructional groups when their rate of progress appears to differ significantly from what was predicted or from the rest of the group's rate, thus keeping achievement and skill levels of the members relatively homogeneous. Thus, in principle, a unified program for low-achieving and LD students would not violate the standard of instructional validity that calls for pacing instruction consistent with student learning rate.

Appropriateness of Instructional Type

In our third question--Is the type of instruction appropriate for one group not also appropriate for the other?--we move from instructional level and rate of learning to a third category of child trait that is potentially relevant to instruction. This can, for lack of a better label, be referred to as "student learning style" that results in the need for a particular brand of instruction.

Experience with the unique instructional needs of people with sensory impairments (e.g., Braille for the blind, sign language for the hearing impaired) invites the generalization that specially adapted instruction is needed by the vast majority of handicapped students (Lloyd, 1984). Moreover, strong intuitive appeal surrounds the notion that different individuals learn in different ways--and, by extension, different teaching methodologies are required for different students. We return to the policy question: Is it instructionally valid to separate categorical programs on the basis of need for different teaching methodologies? Or the converse: Are the instructional approaches appropriate for one special categorical group not equally appropriate for another? We could locate no studies in which categorical label interacted with instructional methodology, nor does there seem to be a literature on this topic (Lloyd, 1984; Morsink, Thomas, & Smith-Davis, 1987).

Even in the most general terms, the concept of aptitude for specialized instruction is long on face validity and short on empirically measured, scientific validity. Attempts to match instruction to learner characteristics such as perceptual abilities (e.g., modality preference) and cognitive abilities (e.g., IQ) have generally met with failure (Arter & Jenkins, 1977, 1979; Kavale & Forness, 1987; Morsink et al., 1987; Ysseldyke, 1973). Given the infrequency of aptitude-treatment interactions, it would be all the more remarkable that gross categorical groupings such as remedial and special education interacted with instructional approaches. Even if educational researchers were to discover strong interactions between treatments and student traits, it is likely that these traits would be distributed throughout students served under existing categorical groupings, resulting in a great overlap in the need for specialized instruction across categories.

As for teacher background and preparation, we hypothesize that the specialized knowledge required for teaching remedial students to read, write, and think quantitatively is essentially the same as that needed to teach mildly handicapped students the same basic skills and vice versa. Data are lacking on this hypothesis, but a study by Marston (1987) is provocative. He reported that within special education, categorically trained teachers (e.g., those teaching EMR, LD, and SED populations) appear to be as successful in teaching students from diagnostic categories that do not correspond with their training as they are in teaching students from "their" diagnostic categories.

Instructional Validity Summarized

We have examined three facets of instructional validity underlying the concept of a noncategorical unified service delivery system for low-achieving students. On each facet--instructional level, learning rate, and need for a particular instructional methodology--the overlap of students across categorical groupings overwhelms any differences in group averages. In the words of Leinhardt et al. (1982), "From an educational point of view, the instructional rationale is the only possible reason for a separate system" (p. 407). The educational problem posed by student diversity arises from different instructional needs of average and high-achieving students versus those of low-achieving students, rather than from differences in etiology or instructional needs among low achievers. Our analysis reveals that the instructional rationale does not support a continuation of separate systems, and that a unified program would be more instructionally valid. In short, current educational policy conflicts with principles of effective instruction.

POLITICAL VALIDITY

We turn now from the classroom to the larger universe where possible solutions to the problem of fragmentation in categorical programs must be addressed. To unify the different programs, participants at all levels--federal, state, and local--must be convinced that the resulting system

will, at the very least, do no harm and that it will bring benefits not achieved under the present system. Stakeholders in the system must have their interests protected; in this case, the stakeholders are the children who receive services, their parents, and professionals--teachers, administrators, and policy makers at all levels--within the system. Political validity comprises three issues: protection and equitable distribution of resources, participation in decision making, and protection of jobs. Basic assumptions underlying these issues are that student rights are guaranteed and that quality services are provided.

Protection and Equitable Distribution of Resources

The major issue for all stakeholders in any changed system is that there is no reduction in services or the budgets that fund services. Protection from funding and service reductions is a concern across the composite of special and remedial programs as well as within individual programs. The goal of unification must not be to economize; rather, it must be to strengthen and stretch available resources so that they are used more efficiently and effectively.

Estimates of the noninstructional costs of compensatory programs vary greatly. Fagan (1987) estimates that 85% Chapter I monies result in instructional services. In contrast, Haggerty et al. (cited in Leinhardt et al., 1982) report that only $400 of every $1,000 spent per child receiving compensatory education results in actual services. Perhaps Haggerty et al. have overestimated the administrative expenditures of compensatory programs; even so, it is evident that program administration and accounting costs are substantial. We suspect that the dilution of financial resources in special education is as great as, if not more than, the dilution in compensatory programs.

Funds saved through a restructuring of categorical programs need to be preserved for services to low-achieving students. They could be used for the following purposes:

1. To amplify instruction routinely provided by special programs (e.g., form smaller, more homogeneous instructional groups).

2. To broaden the scope of programs to include nonroutine services that are both instructional (e.g., teach study skills or social skills) and non-instructional (e.g., provide counseling, social work assistance).

3. To extend services to more students (e.g., include low-performing students in the program who would otherwise be excluded because of limited resources, that is, those students who are said to "fall through the cracks" created by discrete categorical programs).

4. To fortify the general education program (e.g., provide direct assistance to classroom teachers through building level consultants, teacher assistance teams, curriculum adaptation, or inservice training).

An issue closely related to the protection of resources is their equitable distribution. Because every special program was created for a specific client group, maintaining the hard-won resources for those client groups is essential. Advocacy groups want to protect their clients' interests and many suspect that changes will dilute services, undermine established eligibility criteria, relinquish entitlements and rights, or prove harmful in some respect. Advocacy groups would not and should not consider supporting a program unification effort if there were a significant chance that their client group would lose resources to competing groups. At the same time, allocation of resources would have to be rationally based on actual services needed rather than on eligibility criteria by which children are assigned to one program or another based on something as instructionally irrelevant as a formula specifying a minimum discrepancy between IQ and achievement. The problem of entitlement versus eligibility--as in entitlement to special education through P.L. 94-142 versus eligibility for Chapter I or remedial assistance programs--would need to be resolved in favor of entitlement. In any unified system, a sine qua non will be that low-achieving children are entitled to the services they need. Federal, state, and local policy must be modified if resources are to be protected and equitably allocated.

Participation in Decision Making

Even if resources were protected and equitably distributed across low-achieving groups, program integration would require stakeholders' participation in decision making. Parents of children with handicaps have grown accustomed to participating in planning for their children's educational program. These rights should be extended to all parents who have an interest in decisions about their children's placement in a special program. Parents and child advocates are not inextricably bound by the convention of labeling and categorizing students, but they will demand continued participation in decision making about their children's education.

Job Protection

For program unification to be worthwhile, it must redirect unnecessarily or ineffectively used resources toward activities effective in achieving educationally desirable outcomes. Besides simplifying school operations, a probable outcome of reorganizing special programs would be the reallocation of some jobs and the redefinition of others. For example, fewer positions for separate programs could result in fewer administrative duties and less complex fiscal accounting. This would translate into fewer school administrators, financial managers, budget analysts, and accountants. Although some resistance can be anticipated, job reallocation from administrative and fiscal areas could be justified if the preserved resources were applied to direct, programmatic, and instructional services.

Reduction in personnel at the service provider level is not acceptable, but some job redefinition may be required. For example, if unifying separate categories simplified the criteria used to determine a student's eligibility for services, then school psychologists could devote less time to psychometric assessments and more time to consultation on classroom behavior and learning problems, to teaching social skills, and to counseling. Individuals who enjoy their role as psychometricians or who are unsure about their ability to provide other services may resist, but this can be overcome through inservice training, better recognition, and increasing satisfaction in the new or changed role. More difficult to overcome would be the resistance that grows from psycholgists' legitimate fear that their jobs will be reduced or eliminated if their psychometric duties were no longer legislated. Additional positions, including consulting teachers, counselors, educational specialists, and trained aides, could also be created with the redeployment of existing resources now being used for administrative and testing functions.

For a unified program to receive support from professionals, some guarantees would be needed for job security. A simple rule to adopt would be that no service providers lose their jobs, although their roles might undergo some changes. Provisions for job retraining would be a necessary element in re-forming special programs. If teacher or administrator certification requirements were to be changed, those affected would need assurance that any training which they require would be available at no cost, and that they could continue in their jobs during the retraining period.

Benefits to be Derived

Even if the three political issues were resolved to the satisfaction of most people within the educational universe, does the unified system seem so ideal as to be outside the realm of feasibility? Perhaps a brief statement of some of the benefits it would yield will suggest the reasons for at least attempting to make it possible.

Looking at this from the point of view of stakeholders, we begin with administrators. It seems almost self-evident that any process for reducing the number of separate fiscal and program regulations that need to be monitored and adhered to would be welcome as long as there are educational benefits to be derived and not merely efficiency. Maintaining one, not several, accounting systems would increase the coherence of administrators' efforts. In a unified system based on children's entitlements rather than eligibility, administrators would have far fewer policing duties and could contribute to educational decisions about how best to provide services. At still another level of administration, it seems clear that principals, with a less fragmented service system for low-achieving children in their buildings, would be able to spend less time and effort on complicated staffing problems, resource allocation, and bookkeeping.

Teachers in categorical programs face a divisive instructional situation when students receive services from a number of different programs and teachers, and when several programs operate within a building. There is a question of who "really knows" the child best; who is ultimately responsible for his or her performance in school; who is providing what in the child's program. Unless there is extremely close cooperation between the programs and staffs, the results can be educational discontinuity for students and frustration for teachers. It seems obvious that if there were a unified system that provided instruction for all low-performing children, regardless of the problem that causes the poor performance, there would be fewer problems of ownership and more truly collaborative efforts to help children. Support teachers and psychologists working with classroom teachers could provide a more comprehensive picture of a child's problems and how to work through them in the classroom. Support teachers working in a unified system would be relieved of arduous record keeping regarding their time and the nature of services they provide to specific children.

We believe that children will benefit from a system in which they are grouped for instruction on the basis of how they are doing in school rather than on categorical lable. Indeed, the present fragmented system may work to the disadvantage of the children it is supposed to benefit. The artificial grouping of children, based on categorical lable, may inadvertently dilute services by producing unnecessary heterogeneity. The teacher faced with a heterogeneous group generally provides less group instruction and assigns more individual work. The problem, however, is that "individualized instruction" can boomerang and become, quite simply, individual seatwork. The child ends up working alone and not receiving teacher attention. Can we measure the amount of lost instruction time in such a situation? Haynes and Jenkins (1986) found that handicapped children sent to special education resource rooms for reading instruction spent 52% of that time in private seatwork and only 25% of the total resource room time actually reading. There is no little irony in the possibility that a system of categorical programs based on presumably different educational needs may in fact work to the detriment of some children whose needs are more similar than different. This is one of the more serious issues policy makers would have to address.

Finally, parents are likely to be receptive to the idea of a unified system that maintains and amplifies services, that dedicates more resources to direct services and less to administrative and record-keeping functions, that simplifies the current bureaucratic maze associated with independent categorical programs, and that gives them a direct role in decision making about their children's programs.

CONCLUSION

We have discussed two major kinds of issues that influence the present system for educating low-achieving and mildly handicapped children in our elementary schools. There is increasing interest in addressing problems within that system, particularly the widely acknowledged problem of fragmented services. If that interest is in fact as widespread as the literature would suggest, these issues must now be addressed at the policy-making level. Data from studies such as the ones we have discussed can provide answers to questions about the instructional validity of the present system. Those answers can yield one set of realities that will eventually drive policy. But policy development will have to be the outcome of another reality--resolutions to the political validity issues--because there is so much at stake for people throughout this decidedly political system. In our view, the instructional validity questions are no longer the critical issues, because they are more readily answerable and to a large extent answered. We can see no justification for separating students by categorical labels, with the possible exception of those served in bilingual programs in which basic skills are taught in a foreign language. Subgrouping of other low-achieving students, if necessary--regardless of categorical label--should be based only on instructional level. We are not the first to call for a unification of categorical programs for low-achieving students. Earlier, Leinhardt et al. (1982) reached the same conclusion based on an entirely different analysis. Now it is time to examine much more closely the political issues and the obstacles to their resolution.

One major source of fragmentation and irrationally allocated resources is the issue of entitlement vs. eligibility. On the whole, children entitled to services, such as those in special education, fare well because they are protected by law. Other children who are merely eligible for services, for example, Chapter I, do not fare as well because eligibility does not necessarily result in receiving services. We illustrate this by pointing out that between 1976 and 1983 the number of children identified as learning disabled increased from 800,000 to 1.8 million while the number of children served by Chapter I decreased by 1.5 million, despite the fact that there are more children below poverty levels now than before (McGill-Franzen, 1987). The inescapable conclusion is that children are being made to fit labels in order to receive entitled services. One possible solution to this problem is simply to recognize, by policy, that any child not performing well in the regular classroom is entitled to receive appropriate special services. If our data are correct in suggesting that the instructional needs of the members of this more broadly defined group of children are not distinct enough to warrant separate categorical programs, there is hardly any logic in fragmenting the programs to serve them.

Finally, we must emphasize that the foregoing analysis is related to, but not synonymous with, the Regular Education Initiative. Unifying services for low-performing students is one of the elements in Will's (1986) proposal. However, we believe that unification of categorical programs could occur within a framework like the dual system of services that presently exists for regular and low-performing students. Alternatively, the entire educational system could be unified, more along the lines proposed by Will and other spokespersons for the REI (Reynolds et al., 1987). Either way, we are confident that such a unification would free valuable resources that could be better utilized in the service of children with special needs.

REFERENCES

Arter, J. A., & Jenkins, J. R. (1977). Examining the benefits and prevalence of modality considerations in special education. Journal of Special Education, 11, 281-298.

Arter, J. A., & Jenkins, J. R. (1979). Differential diagnosis-prescriptive teaching: A critical appraisal. Review of Educational Research, 49, 517-555.

Birman, B. F. (1979). Case studies in overlap between Title I and P.L. 94-142 services for handicapped students. U.S. Department of Health, Education and Welfare Research Report EPRC 26. Menlo Park, CA: SRI International.

Candler, A. C., Johnson, N. L., & Green, C. (1983). The differences among children with learning problems. Education, 104, 219-223.

Deno, S. L., Mirkin, P. K., & Chiang, B. (1982). Identifying valid measures of reading. Exceptional Children, 49(1), 36-45.

Fagan, T. (1987). Personal communication.

Felix, N., Hertlein, J. C., McKenna, D., & Rayborn, R. (1987). Combining categorical program services can make a major difference. KAPPAN, 68(10), 787-788.

Fuchs, L., & Deno, S. L. (1981). A comparison of reading placements based on teacher judgment, standardized testing, and curriculum-based assessment. (Research Report No. 56) Minneapolis: University of Minnesota, Institute of Research on Learning Disabilities.

Haynes, M. C. & Jenkins, J. (1986). Reading instruction in special education resource rooms. American Educational Research Journal, 23(2), 161-190.

Jenkins, J. R., Peterson, D. L., & Schrag, J. (1988, in preparation). Similarities and differences between remedial and special education students. Seattle: Experimental Education Unit, University of Washington.

Kavale, K. A., & Forness, S. R. (1987). Substance over style: Assessing the efficacy of modality testing and teaching. Exceptional Children, 54(3), 228-239.

Leinhardt, G., Bickel, W., & Pallay, A. (1982). Unlabeled but still entitled: Toward more effective remediation. Teachers College Record, 84(2), 391-422.

Lloyd, J. W. (1984). How should we individualize instruction, or should we? Remedial and Special Education, 5(1), 7-15.

Marston, D. (1987). Does categorical teacher certification benefit the mildly handicapped child? Exceptional Children, 53(5), 423-431.

Marston, D. & Magnusson, D. (1985). Implementing curriculum based measurement in special and regular education settings. Exceptional Children, 52(3), 266-276.

McGill-Franzen, A. (1987). Failure to learn to read: Formulating a policy problem. Reading Research Quarterly, 22(4), 475-490.

Morsink, C. V., Thomas, C., & Smith-Davis, J. (1987). Noncategorical programs: Process and outcomes. In M. C. Wang, M. C. Reynolds, & H. J. Walberg (Eds.), The handbook of special education: Research and practice (pp. 287-312). Oxford, England: Pergamon Press.

O'Shea, L. J., & Vacante, G. (1986). A comparison over time of relative discrepancy scores of low achievers. Exceptional Children, 53(3), 253-259.

Reynolds, M. C. & Wang, M. C. (1983). Restructuring "special" school programs: A position paper. Policy Studies Review, 2(1), 189-212.

Reynolds, M. C., Wang, M. C., & Walberg, H. J. (1987). The necessary restructuring of special and regular education. Exceptional Children, 53(5), 391-398.

Shinn, M. R., Tindal, G. A., Spira, D., & Marston, D. (1987). Practice of learning disabilities as social policy. Learning Disability Quarterly, 10(1), 17-28.

Shinn, M. R., Ysseldyke, J. E., Deno, S. L., & Tindal, G. A. (1986). A comparison of differences between students labeled learning disabled and low achieving on measures of classroom performance. Journal of Learning Disabilities, 19(9), 545-552.

Slavin, R. E. (1987a). A theory of school and classroom organization. Educational Psychology, 22(2), 89-108.

Slavin, R. E. (1987b). Grouping for instruction in the elementary school. Educational Psychologist, 22(2), 109-127.

Stainback, W., & Stainback, S. (1984). A rationale for the merger of special and regular education. Exceptional Children, 51(2), 102-111.

Will, M. (1986). Educating students with learning problems--a shared responsibility. Washington, DC: Office of Special Education and Rehabilitation Services, U.S. Department of Education.

Ysseldyke, J. E. (1973). Diagnostic-prescriptive teaching: The search for aptitude-treatment interactions. In L. Mann & D. Sabatino (Eds.) The first review of special education. Philadelphia: JSE Press.

Ysseldyke, J. E., Algozzine, B., Shinn, M. R., & McGue, A. (1982). Similarities and differences between low achievers and students classified as learning disabled. Journal of Special Education, 16, 73-85.

JOSEPH R. JENKINS is Professor of Special Education, University of Washington, and Director of the University's Experimental Education Unit. CONSTANCE G. PIOUS is the Coordinator of Publications at the Experimental Education Unit. DAVID L. PETERSON was Coordinator of the Enhancing Instructional Program Options project.
COPYRIGHT 1988 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1988 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Jenkins, Joseph R.; Pious, Constance G.; Peterson, David L.
Publication:Exceptional Children
Date:Oct 1, 1988
Words:7349
Previous Article:Response to Wang and Walberg.
Next Article:Formulating optimal state early childhood intervention policies.
Topics:


Related Articles
Evaluation of the adaptive learning environments model.
Four fallacies of segregationism.
What's happening in self-contained special education classrooms?
State variation in placement of children with handicaps in segregated environments.
Students' time on learning tasks in special education.
Educating all children: ten years later.
Defining mild disabilities with language-minority students.
A response to Esposito and Koorland: a bias in search of supporting data.
Special education and the regular education initiative: basic assumptions.
Counterpoint: "20/20 Analysis" - Phase 1 only.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters