Printer Friendly

Caseload in Special Education: An Integration of Research Findings.

Optimal class size in general education has been studied extensively in past decades, prompting analysis of the issue from many perspectives and acclamation of small classes as a panacea for a myriad of school problems (Goldstein & Blatchford, 1998). Class size reductions have become so widely supported, in fact, that fully half of our 50 states are currently implementing some type of class size reduction (Wexler et al., 1998), and a proposal for federally-funded class size reductions is among the 1998 education goals (National Education Goals Panel, 1998). Although class size research is replete with challenges and disparate conclusions, scholarly endeavors focusing on the relationship between general education class size and student achievement abound, providing a wealth of information for stakeholders and policymakers (Folger & Breda, 1989; McRobbie, Finn, & Harman, 1998; Odden, 1990; Pate-Bain, Fulton, & Boyd-Zaharias, 1999).

Research into caseload in special education, on the other hand, has been conducted far less frequently. With the first focused studies on special education caseload beginning little more than a decade ago (Ysseldyke, 1988), and few major research projects occurring since, relatively little is known about the impact of caseload on special education student performance. Much of the extant research has based its conclusions on the relationship between group size and student engagement rather than on a direct link between caseload and achievement, further limiting the availability of empirical information upon which to base policy. Using a regression equation, Algozzine, Hendrickson, Gable, and White (1993) predicted a significant decrease in student achievement when special education caseloads were increased. Empirical evidence, however, has yet to validate this hypothetical prediction. In her analysis of special education guidelines, O'Hearn (1995) concurred with this limitation, stating that "special education policymakers will not find enough specific resources on caseload/class size to assist in revising caseload/class size requirements" (p. 5).

Notwithstanding the limited availability of research findings, the issue of caseload in special education has taken on an increasingly significant role in our education system. Special education costs spiral ever higher, threatening to impinge upon general education budgets in some districts. The number of students participating in special education has increased by 47% between 1977 and 1995, compared with a concurrent 2% decrease in general education enrollment (National Center for Education Statistics, 1997). Many special educators reported that increases in caseloads corresponded with simultaneous increases in meetings and paperwork demands. Eighty-seven percent of special educators, in fact, reported that they did not have enough time to spend with each individual student during the 1998-99 school year (The Council for Exceptional Children, 1998). Accumulated evidence suggests caseload is one factor in teacher attrition, which continues to exceed that of general educators, exacerbating the problem of staff shortages in special education.

Despite the escalating costs of special education, the ever-growing number of students needing services, and the reported time constraints of special educators, policies governing caseload and class size have remained inconsistent across the 50 states (Rylance, Chiang, Russ, & Dobbe-Whitcomb, 1999). Statewide caseload guidelines range from complex formulas considering settings, disability category, paperwork, and severity of disability to policies requiring that all decisions be made by local school districts (Rylance et al.). Implementation of the existing caseload policies has been even more variable. A nationwide analysis of caseloads for teachers of students who are severely emotionally/behaviorally disturbed found state averages ranged from 3 to 35 students per teacher (Algozzine et. al., 1993). A 1999 Council for Exceptional Children (CEC) survey of special educators revealed caseloads ranging from 4 to 154 students per teacher. Both studies, however, found similar average caseloads for the groups considered--12 and 14 students per teacher respectively--lending credence to the claim that smaller class size is the main reason for placing special education students in more restrictive settings (Ysseldkye & Algozzine, 1995).

Studies related to special education caseload, instructional group size, and teacher attrition compose the bulk of this review. Articles were located using Internet browsers and academic databases, including Academic Search, Master Files, ERIC, and Wilson Web using keywords "caseload," "class size," "pupil-teacher ratio," and related synonyms. References from relevant articles provided additional sources for research studies. The paucity of research studies related to caseload or instructional groupings prohibited the use of explicit criteria for including articles in this integration. Although literature related to teacher attrition was far more readily available, only studies identifying caseload as a dependent or independent variable warranted inclusion. Rather than relying on estimates of effect size necessary to conduct a meta-analysis of special education caseload, this microanalytic approach is intended to summarize and analyze the few existing studies as cases supporting a common practice. Table 1 summarizes the primary studies addressed herein.

TABLE 1 Studies Relating Class Size to Student Engagement and Achievement
Study N Method

Fink & Sandall, 1978 12 students Pretest/Posttest
Forness & Kavale, 1985 393 students Trained observation
Gottlieb & Alter, 1997 45 schools Trained observation,
 Interviews, Data reviews
Keith, Fortune, & Keith, 722 students Teacher surveys, Trained
1993 observations, Record
McCabe, Jenkins, Mills, 24 students Transcribed videotapes,
Dale, Cole, & Pepler, in which utterances,
1996 different words, and
 length of statements
 were divided by
 number of seconds in
Snart & Hillyard, 1985 15 teachers Trained observation
Thurlow, Ysseldyke, 139 students Trained observation,
Wotruba, & Algozzine, CISSAR, & IFS
Wesding, Ferrell, & 9 students Trained observation of
Swenson, 1982 individual children;
 review of individual
Wheeler, 1993 433 teachers Teacher survey, Burnout

Study Subjects

Fink & Sandall, 1978 Preschoolers with
 Down syndrome, CP
Forness & Kavale, 1985 Children with mild
 MR; Mean age 11.6
Gottlieb & Alter, 1997 Children with mild

Keith, Fortune, & Keith, Children with LD,
1993 ED, MR
McCabe, Jenkins, Mills, Preschool children
Dale, Cole, & Pepler, with language delays
1996 and various other
Snart& Hillyard, 1985 6 18-year-olds with
 profound disabilities
Thurlow, Ysseldyke, Elementary children
Wotruba, & Algozzine, with mild LD, EBD,
1993 EMR
Wesding, Ferrell, & Children aged 3-9
Swenson, 1982 with severe or pro-
 found retardation
Wheeler, 1993 LD, MR, ED, Mild-

Study Group Sizes

Fink & Sandall, 1978 1:1, 12:1 integrated
Forness & Kavale, 1985 12, 15, and 19 pupils/
Gottlieb & Alter, 1997 5:1 and 8:1; Resource
Keith, Fortune, & Keith, Not clearly indicated
McCabe, Jenkins, Mills, Play groups with 2 or
Dale, Cole, & Pepler, 4 students
Snart& Hillyard, 1985 10:3, 8:3, 7:3
Thurlow, Ysseldyke, 1:1, 3:1, 6:1, 9:1, 12:1
Wotruba, & Algozzine,
Wesding, Ferrell, & 3:3 for 30 min, 1:1
Swenson, 1982 for 10 min
Wheeler, 1993 Head count 14;
 Median FTE 3

Study Major Findings

Fink & Sandall, 1978 * Minimal increases in reading under
 1:1 tutorial
 * Minimal decreases in math under 1:1
Forness & Kavale, 1985 * Communication with peer or teacher
 increased in smaller classes
 * Attending behavior decreased in
 smaller classes
 * Disruptions decreased in smaller
Gottlieb & Alter, 1997 * Reading test scores increased in
 smaller classes
Keith, Fortune, & Keith, * Increases in reading in smaller groups
1993 * Increases in math in smaller groups
 * Greatest increases in elementary
McCabe, Jenkins, Mills, * Increased number of utterances in
Dale, Cole, & Pepler, pairs
1996 * Increased usage of different words and
 length of utterances in quartets
Snart& Hillyard, 1985 * Instructional time rose from 12.6% to
 23.2% in student-teacher groupings of
 * Noninstructional [time.sup.a]
Thurlow, Ysseldyke, * ART (1:1, 3:1 > all others)
Wotruba, & Algozzine, * AET (1:1 > all others)
1993 * Inappropriate behavior (12:1 > 1:1)
 * Task management responses (12:1 > 3:1)
Wesding, Ferrell, & * Increased number of instructional rain
Swenson, 1982 teaching behavioral objectives from
 IEP (1:1 > 3:3)
 * Decreased number of [turnaways.sup.b]
 per number of min of instruction (1:1
 < 3:3)
Wheeler, 1993 * Teachers perceive students' academic
 and social/behavioral achievement to
 be greater when caseloads are smaller

(a) Noninstructional time: Time spent on toileting, moving to and from class, and other matters not related to IEP instruction; (b) Turnaways: The number of times a teacher directs his or her attention away from students he or she is instructing CP = cerebral palsy;, MR = mental retardation; LD = learning disabilities; ED = emotional disturbances; EBD = emotional-behavioral disorders; EMR = educable mental retardation; ART = academic responding time (includes writing, playing games, reading silently, asking and answering questions); ADT = academic engagement time (academic responding time combined with passively attending to instruction).

In order to analyze the elusive relationship between caseload and achievement within the parameters of available research, findings are synthesized to consider the links between (a) group size and student engagement, (b) group size and academic achievement, and (c) caseload and special education teacher attrition. Conclusions and research recommendations complete the review.


A number of studies beginning in the 1970s analyzed the link between special education instructional group size and student engagement (Kamps & Walker, 1990; Keith, Fortune, & Keith, 1993; Ranieri, Ford, Vincent, & Brown, 1982; Thurlow, Ysseldyke, Wotruba, & Algozzine, 1993; Westling, Ferrell, & Swenson, 1982). Because academic engaged time has corresponded directly with academic achievement for students with mild, moderate, or severe cognitive disabilities (Kamps & Walker, 1990; Logan & Keefe, 1997), improvement in engaged time may be viewed as an indicator of instructional efficacy. The following paragraphs consider the engagement efficacy of instructional group size for students with mild disabilities and those with severe disabilities.


Overall, engagement for students with mild disabilities correlated inversely with the number of students in a group (Forness & Kavale, 1985; Keith et al., 1993; McCabe et al., 1996; Thurlow et al., 1993). The largest of these studies, involving 139 students and 54 teachers, revealed an incremental decline in academic response and engagement times in larger group sizes (Thurlow et al.). These differences reached statistical significance between groups of 1 to 3 students and those with 6 to 12 students for response time and between groups with 1:1 student-teacher ratios and all larger groups for engagement time. At a similar rate, inappropriate behaviors occurred more frequently in higher group sizes, with the exception of the 9:1 grouping, in which such behaviors occurred slightly less often than in 3:1 ratios.

Keith et al. (1993) investigated the impact of class size and class mix on the academic outcomes of 139 Virginia students with varying disabilities in preschool through Grade 12. Comparison groups included special education classrooms with approved waivers allowing classes to exceed state mandated maximums (larger groupings) and classes operating without waivers (smaller groupings). Statewide maximums ranged from 6 to 24 students based upon type of disability, setting, and presence of a paraprofessional. Significantly higher rates of appropriate behavior were noted in smaller class groupings, although no significant differences were revealed when class mix was considered.

In a comparison of 12, 15, and 19 pupils with mild mental retardation, Forness and Kavale (1985) noted increases in attentive behavior in larger classrooms, but more frequent communicative responses and fewer disruptive actions in smaller classes. Through surveys, special education teachers further substantiated these findings, revealing their perceptions that student social interactions and behavioral appropriateness improved when caseloads/group sizes were smaller (Wheeler, 1993; Ysseldyke, 1988). When asked about mixing students with different disabilities within a classrooms, however, teachers differed from administrators in their overall opinion that students should not be mixed (61% and 16%, respectively).

For several reasons, these findings may not necessarily be generalized to contemporary educational settings. First, the studies took place more than a decade ago, during which time significant societal, educational, and legal changes transpired. The resulting instructional setting may be sufficiently different to minimize the pertinence of earlier findings. Second, the studies seemed merely to confirm what may be viewed as a logical surmise: Students in small groups have more opportunities to participate, resulting in an inevitable increase in the frequency of participation.

Finally, and perhaps most significant, each of the studies involved multiple educators and a range of academic subjects. While group size may indeed have impacted student engagement, it is plausible that the quality of the teacher, student interest and motivation, or the nature of the subject may have exerted a greater impact. The anomalistic finding by Forness and Kavale (1985) indicating lower attentive behavior in smaller classes, for instance, may have occurred because the smaller groups focused on routine skills such as phonics or basic arithmetic computation. It is equally conceivable that class sizes were larger in subjects highly appealing to children, such as those involving tactual participation or intriguing discussions.


The optimal group size for many aspects of student engagement was 1:1 (Fink & Sandall, 1980; Kamps & Walker, 1990; Ranieri et al., 1982). In 1:1 instruction, students engaged themselves actively in academic work for a significantly greater portion of time than in larger groups. At the same time, decreases in inappropriate student behavior allowed teachers to devote more time to instruction and less time to task and behavior management. Similar but less consistent trends occurred when group sizes were increased to 2:1 or 3:1.

These findings diminished in significance, however, when inclusive time spans rather than only instructional min were sampled. Ranieri et al. (1982) noted that high rates of on-task behavior during instruction were negated when the resulting free time for all students was included in the analysis. In other words, the number of inappropriate behaviors occurring when the teacher provided 1:1 instruction to other students was high enough to offset the positive engagement during instruction.

When paraprofessionals were present in an instructional setting, the groupings of students for instructional purposes influenced both instructional time and student engagement. In a setting consisting of one teacher, two aides, and nine students with profound mental retardation, Westling, et al. (1982) observed increased teacher attention and fewer turnaways in 10 minutes of 1:1 instruction than in 30 min of 3:1 instruction. Snart and Hillyard (1985) also observed that students with multiple disabilities assigned to a teacher in groups of two for shorter periods of time received a greater amount of instructional time than those working in larger groups with teacher/aide combinations for longer periods of time.

Although findings tend to support 1:1 instruction as a viable grouping for increasing student engagement, many drawbacks exist that may mitigate its effects. First, an individual instructional model may inadvertently nurture student dependence on the instructor, inhibiting development of independence and autonomy. Second, 1:1 instruction limits opportunities for students to learn vicariously from their peers and develop the social skills necessary to succeed outside the classroom. Third, 1:1 instruction demands extended teacher attention and necessitates substantial free time for those students not involved with the teacher, as indicated by Ranieri et al. (1982). In a population with severe disabilities, such free time is often accompanied by negative behaviors, discounting the positive engagement impact of the individual instruction.

In summary, the results of earlier studies relating special education instructional group size and student engagement demonstrated with reasonable consistency that engagement for students with mild disabilities increased when group size decreased, regardless of age or type of disability. Furthermore, such improvements in student engagement occurred despite occasional decreases in actual time available for instruction. Finally, some benefits of small group instruction outweigh those of 1:1 instruction, ameliorating its necessity as a primary instructional model.


Measuring academic achievement for special education students presents a formidable challenge. For a widely diverse population in which educational and assessment objectives retain individual components, obtaining valid, comparable measures remains fraught with difficulties. Unlike general education students, many students with disabilities need not participate in statewide testing. For those who do participate, the test results may not reflect individualized education program (IEP) objectives and, thus, may not accurately reflect progress. When considering instructional group size as a variable, the issue further multiplies in complexity due to variations in special education groupings and settings not present in relatively stable general education class sizes.

Despite the research obstacles relating instructional group size to student achievement, several studies addressed this issue (Gottlieb & Alter, 1997; Keith et al., 1993; McCabe et al., 1996; Thurlow et al., 1993). To measure academic achievement for students with mild disabilities, researchers utilized standardized test scores, task completion, and task success rates. Studies involving students with severe disabilities relied upon pre- and posttests related to specific educational objectives to determine academic achievement in various groupings. Academic findings were in many instances supplemented by surveys or interviews. The following paragraphs consider the impact of instructional group size on (a) students with mild disabilities, (b) students with severe disabilities, and (c) instructional practices.


Two large-scale studies investigating the impact of instructional group size and academic achievement drew essentially the same conclusion: Larger group sizes correlated inversely with academic achievement for special education students. Gottlieb and Alter (1997) based this conclusion on their evaluation of mandated increases from five to eight students in New York City resource and speech language classrooms. Results from statewide reading achievement tests revealed that only 16% of sixth graders met state reading criteria after group size increases, compared with 29% before increases (1994-95). Test scores of 3rd- and 8th-grade students also dropped, resulting in respective declines of 5.4% and 9.2% in the number of students with learning disabilities meeting state requirements for reading goals. Based on these findings, Gottlieb and Alter recommended that "no more than five students receive resource room instruction at any one time" (p. 3). Because program evaluation began after group size increases had already been initiated, however, this study could not compare other relevant academic variables before and after increases.

Findings from Keith et al. (1993) revealed that IEP progress and academic achievement in reading, math, science, and social studies reached significantly higher rates in smaller classes, with the greatest impact occurring in reading and with elementary students. Additionally, they found that students in single-disability classes "achieve at a higher level in reading, math, and social studies" than those in mixed classes (p. 66). These findings are limited, however, by the absence of detailed information related to actual group sizes and severity of disabilities of the students involved.

Despite commonly held confidence in 1:1 instruction as an ideal model, few studies exist examining differences in academic achievement between students with mild disabilities receiving 1:1 instruction and those receiving instruction in small groups. In one such study, Thurlow et al. (1993) discovered similar rates of task completion and task success in students with mild disabilities in all group sizes. A ceiling effect, however, mitigated the impact of the study's results for 1:1 instructional groupings. As revealed in a study by Torgeson, Wagner, Rashotte Alexander, & Conway (1997), students with mild disabilities receiving 80 hours of 1:1 instruction achieved high academic gains, while students receiving instruction in resource rooms with up to 15 students achieved lower gains.

Surveys related to two of the above studies supported findings that smaller group sizes prompted greater academic achievement. Eighty-three percent of teachers interviewed by Gottlieb and Alter (1997) indicated that an increased instructional group size made it "more difficult for them to deliver effective instruction ... remediate children's educational deficits ... plan consistent educational plans ... and offer individual instruction" (p. 19). Virginia special education teachers and directors believed a 24-student caseload maximum for resource rooms was too high although both groups accepted the 8-student caseload maximum in self-contained classrooms.

These findings suggest that lower group sizes facilitate greater academic achievement for students with mild disabilities. Insufficient evidence exists to support or refute the efficacy of 1:1 instruction on academic achievement of students with mild disabilities, although a federal mandate demands individualized education programs within the least restrictive environment.


For students with severe disabilities, small group instruction was found to be effective in providing academic achievement. Although findings remained somewhat inconsistent, many studies found that academic achievement under small group instruction equaled achievement under 1:1 sessions. (Favell, Favell, & McGimsey, 1978; Fink & Sandall, 1978; Fink & Sandall, 1980; Kamps & Walker, 1990; Ranieri et al., 1982; Westling et al., 1982).

As with students with mild disabilities, academic achievement of students with severe disabilities was similar for students in 1:1 and larger groupings ranging from 3:1 to 6:1. In verbal recall activities including four preschool children with developmental delays, Fink and Sandall (1978, 1980) found no significant differences on immediate or delayed word recognition tests between those instructed 1:1 and those instructed in small groups. In a second study, Fink & Sandall (1980) obtained similarly insignificant results, with reading scores in 1:1 sessions revealing slight increases and arithmetic scores slightly decreasing. Studying three children with autism, Kamps & Walker (1990) found that no significant differences occurred between teacher instruction in 1:1 or 3:1 ratios. Finally, examining eight children with mental retardation, Westling et al. (1982) found no significant differences in the number of objectives taught nor in the number of correct responses.

Although findings within these studies are consistent, each presumes that all students within a group require similar instructional techniques and comparable learning objectives. If this is indeed the case, as it may be for students with similar severe disabilities, then small group instruction is clearly equally effective and more time-efficient than 1:1 instruction. If instructional needs differ between students, the results of these studies may not apply.

To summarize, although 1:1 instruction revealed increased participation of students with severe disabilities, it has not produced significantly greater academic gains. Furthermore, the potentially negative effects of 1:1 instruction, such as nurturing dependency, may serve to counter the gains in student engagement. As a component of an instructional program, 1:1 grouping offers clear benefits. As a dominant educational model, however, 1:1 instruction appears to be neither cost-effective, time efficient, nor pedagogically justified.


A desired outcome of class size reductions in general education rests on a hypothesized increase in individualized instruction. For the same reason, many educators not only deem small special education class sizes beneficial but essential. As in general education studies, however, researchers observing special education classrooms discovered few differences in teaching practices even when class sizes varied. In fact, group sizes frequently needed to reach 1:1 ratios before teachers distinctly individualized their instruction (Slavin, 1990).

Gottlieb and Alter (1997) found that resource room teachers with class sizes of up to 8 students spent 61% of time on whole group instruction compared with only 3% on individualized instruction, but comparison data from the smaller classes in earlier years were unavailable. Keith et al. (1993) found that large group instruction increased as class sizes increased while individualization, infrequent in any of the classes observed, occurred slightly more often in smaller classes. O'Hearn (1995) reported that less time centered on academic instruction and more time on student behavior in larger classes, with the greatest differences occurring between classes smaller than 9 students and those greater than 13. Minimal differences were found in instructional planning, practice, and presentation in groupings ranging from 1:1 to 12:1, although the number of times teachers checked with students for understanding were significantly greater in smaller than larger groups (Thurlow et al., 1993).

Class size exerted little or no impact on the variety of teaching methods used, but teachers in cross-categorical classrooms used significantly fewer teaching methods than did those in single-disability classrooms. Studying six special education teachers in resource room settings, Vaughn, Moody, and Schumm (1998) found that whole group reading instruction was dearly the norm, with little variation in instruction or materials between students. Further, teachers perceived that few differences exist between instructional methods for general and special education students.

Such findings seem to indicate that individual teachers rather than class size alone determine the use and frequency of optimal teaching practices. While such individualized instruction benefits student achievement in general education, it is essential for student success in special education. The IEP itself, in which teachers determine unique long- and short-term objectives for each student, encapsulates this premise. The lack of correlation between individualized instruction and caseload presents an issue for concern, not only by caseload policymakers, but also by staff development planners, teacher education programs, and special educators themselves.


Although figures vary, large numbers of teachers have left education to seek jobs elsewhere. Louis Harris and Associates (1988) found that 60/0 to 80/0 of general educators turned from teaching to other professions. The attrition rate among special educators has appeared considerably higher than their general education counterparts (Billingsley, 1993; Lauritzen, 1997). Singer (1993) found 10% of all special educators left teaching and revealed that the average special educator abandoned education as a profession within 6 years of beginning teaching.

Many researchers suspect that large caseloads contributed to the high attrition rate among special educators (Wisniewski & Gargiulo, 1997). Indeed, CEC (1999) announced a major effort to study adverse teaching conditions. Foremost among the variables to be studied were unmanageable caseloads. In support of their research agenda and based on a recent survey, CEC revealed that 61% of special education teachers cited large caseloads and class sizes as a major problem (Sack, 1998). Another survey study revealed that teachers left education due to perceptions of poor school climate including caseload size (Miller, Brownell, & Smith, 1999).

Although many studies have been conducted to identify causes of high teacher attrition, few have considered caseload as an independent variable. While professionals and researchers suspect that caseload and class size exerted a powerful influence on special education attrition rates, little empirical evidence currently exists. Nevertheless, the impact of caseload on teacher attrition warrants serious consideration, for continually high rates of teacher attrition create teacher shortages, forcing districts to hire unqualified individuals or increase caseloads for existing staff. Either condition leads to diminution of service quality for children in need of precise, highly skilled, and specialized instruction.


Evidence throughout the existing research supports the belief that lower instructional group sizes are essential to both student engagement and achievement. Larger caseloads naturally increase group sizes and minimize opportunities for individualization and academic success. Special educators studied by Moody, Vaughn, Hughes, and Fischer, 2000) epitomized the linkage between caseload and group size, revealing that their "caseloads made it extremely difficult to provide students with individualization" (p. 314). At the same time, no identifiable caseload practice has consistently produced positive outcomes for students with disabilities. Before research can guide policymakers, several questions require further empirical study as well as practical consideration.

First, how can completion of IEP goals and objectives be used to assess caseload and group size efficacy? The limited applicability of standardized tests as measurement tools for children with diverse special educational needs demands use of a more individually applicable tool. At the same time, federal law mandates IEP goals for each identified student. Thorough examination of these goals would serve as an ability-appropriate assessment tool for both research and evaluation purposes.

Second, how does the current trend toward inclusion and cross-categorical instruction impact optimal caseload? As more students in special education enter general education settings, it becomes essential for special and general educators to effectively collaborate with other professionals in their own and other specialties. Additionally, effective and efficient use of teacher and student time becomes a challenge and goal. Time allocations, materials sharing, ownership problems, and collaboration warrant careful consideration.

Third, in what ways can special educators enhance their skills in individualizing instruction within a small group setting? Clearly a benefit and necessity to special education, individualization must occur to ensure effective instruction. Nevertheless, studies revealed that individualization does not generally occur until student-teacher ratios reach 1:1, impractical for students with mild disabilities. Effective strategies utilizing individualized instruction within group settings must be identified, and teachers must be guided and supported in daily implementation of such strategies.

Fourth, what differences exist in the impact of caseload on special education students in rural versus urban areas? Frequently, rural school districts with few special education students must compromise the student's, the teacher's, or the school's best interests to facilitate service provision. Necessary differences in caseload policies, if any, between rural and urban areas present a worthwhile topic of further study.

Finally, in what ways do general education policies and procedures impact special education caseloads? Numerous initiatives in general education impact special education caseload capacity and student needs. General education class size reductions, for instance, may create a climate in which the general education teacher can implement individualization more frequently. Programs such as Reading Recovery may provide specialized instruction in reading and language arts to minimize needs for special education services. In addition, the presence of general education aides and volunteers may diminish the need for special education paraprofessionals in inclusionary classrooms with fewer students. Patently, class size and practices in general education exert a huge impact on caseloads and practices in special education.

The extant research provides few clear empirical directions for polio/makers, administrators, and educators attempting to formulate consistent caseload policies. A myriad of complicating factors, which include inclusionary settings, cross-categorical models, and IDEA reauthorization, steer a complex problem into still murkier waters. Several elements, however, emerge as worthy of consideration in the development of caseload policy. First, smaller class sizes promote higher levels of engagement and instructional individualization. Second, smaller class sizes suggest an association with higher teacher retention rates. Third, caseload size needs to account for the severity of disability and the age of the students. Finally and perhaps most important, the development of caseload determinations and policies demands consideration of the interface between general and special education, as well as the synergistic dynamics within the school.


Algozzine, B., Hendrickson, J. M., Gable, R. A., & White, R. (1993). Caseloads of teachers of students with behavioral disorders. Behavioral Disorders, 18, 103-109.

Billingsley B. S. (1993). Teacher retention and attrition in special and general education: A critical review of the literature. The Journal of Special Education, 27, 137-174.

The Council for Exceptional Children. (1998, February). CEC launches initiative on special education teaching conditions. CPC Today Online [Serial]. Available: setc.htm [1999, Mar. 4].

The Council for Exceptional Children (1999, June/July). Special educators share their thoughts on special education teaching conditions. CEC Today, 5 (9), 1, 5.

Favell, J. E., Favell, J. E., & McGimsey, J. F. (1978). Relative effectiveness and efficiency of group versus individual training of severely retarded persons. American Journal of Mental Deficiency, 83, 104-109.

Fink, W., & Sandall, S. (1978). One-to-one versus group academic instruction with handicapped and nonhandicapped preschool children. Mental Retardation, 16, 236-240.

Fink, W. T., & Sandall, S. R. (1980). A comparison of one-to-one and small group instructional strategies with developmentally disabled preschoolers. Mental Retardation, 18, 35-36.

Folger, J., & Breda, C. (1989). Evidence from Project STAR about class size and student achievement. Peabody Journal of Education, 76(1), 17-33.

Forness, S. R., & Kavale, K. A. (1985). Effects of class size on attention, communication, and disruption of mildly mentally retarded children. American Educational Research Journal, 22, 403-412.

Goldstein, H., & Blatchford, P. (1998). Class size and educational achievement: A review of methodology with particular reference to study design. British Educational Research JournaL 24, 225-239.

Gottlieb, J., & Alter, M. (1997). An evaluation study of the impact of modifying instructional group sizes in resource rooms and related service groups in New York City. New York: New York University. (ERIC Document Reproduction Service No. ED 414 373)

Kamps, D., & Walker, D. (1990). A comparison of instructional arrangements for children with autism served in a public school setting. Education and Treatment of Children, 13, 197-216.

Keith, T. Z., Fortune, J. C., & Keith, P. B. (1993). Special education program standards study, Commonwealth of Virginia (Final Technical Report). Richmond: Virginia State Department of Education. (ERIC Document Reproduction Service No. ED 369 207)

Lauritzen, P. (1997). Supply and demand of educational personnel for Wisconsin public schools. An examination of data trends, 1997. Madison: Wisconsin State Department of Public Instruction (ZQU9875). (ERIC Document Reproduction Service No. ED 415 211)

Logan, K. R., & Keefe, E. B. (1997). A comparison of instructional context, teacher behavior and engaged behavior for students with severe disabilities in general education and self-contained elementary classrooms. The Journal of the Association for Persons with Severe Disabilities, 22, 16-27.

Louis Harris & Associates (1988). The metropolitan life survey of American teachers: Strengthening the relationships between teachers and students. New York: Metropolitan Life.(*)

McCabe, J. R., Jenkins, J. R., Mills, P. E., Dale, P. S., Cole, K. N., & Pepler, L. (1996). Effects of play group variables on language use by preschool children with disabilities. Journal of Early Intervention, 20, 329-340.

McRobbie, J., Finn, J. D., & Harman, P. (1998). Class size reduction: Lessons learned from experience. Policy Brief No. 23 [On-line]. San Francisco: WestEd. Available: [1999, June 26]. (ERIC Document Reproduction Service No. ED 423 600)

Miller, M. D., Brownell, M. T., & Smith, S. W. (1999). Factors that predict teachers staying in, leaving, or transferring from the special education classroom. Exceptional Children, 65, 201-218.

Moody, S. W., Vaughn, S., Hughes, M. T., & Fischer, M. (2000). Reading instruction in the resource room: Set up for failure. Exceptional Children, 66, 305-316.

National Center for Education Statistics. (1997). The condition of education, education of students with disabilities [On-line]. Available: nces/pubs/ce/c9746a0l.html [1999, Aug. 19].

National Education Goals Panel (1998, Feb. 4). Daily Report Card [On-line]. Available: [1999, July 7].

Odden, A. (1990). Class size and student achievement: Research-based policy alternatives. Educational Evaluation and Policy Analysis, 12, 213-227.

O'Hearn, E. (1995). Caseload/class size in special education: A brief analysis of state regulations (Contract No. HS92015001). Alexandria, VA: National Association of State Directors of Special Education. (ERIC Document Reproduction Service No. ED 390 193)

Pate-Bain, H., Fulton, B. D., & Boyd-Zaharias, J. (1999). Effects of class-size reduction in the early grades (K-4) on high school performance [On-line]. Health and Education Research Operative Services, (HEROS) Inc. Available: www.telalink.nit/~heros [1999, July 15].

Ranieri, L., Ford, A., Vincent, L., & Brown, L. (1982). The relative effectiveness of one-to-one versus one-to-three instructional arrangements with severely multihandicapped students. Madison: University of Wisconsin-Madison and Madison Metropolitan School District. (ERIC Document Reproduction Service No. ED 162 122)

Rylance, B. J., Chiang, B., Russ, S., & Dobbe-Whitcomb, S. (1999, May). Special education caseload and class size policies in the fifty states. Madison: Wisconsin State Department of Public Instruction.(*)

Sack, J. L. (1998, April 29). Special educators fault workload. Education Week on the Web [On-line]. Available: ew/1998/33cec.h17 [1998, Oct. 12].

Singer, J. (1993). Are special educators' career paths special? Results from a 13-year longitudinal study. Exceptional Children, 59, 262-279.

Slavin, R. E. (1990). Class size and student achievement: Is smaller better? Contemporary Education, 62(1), 6-12.

Snart, F., & Hillyard, A. (1985). Staff ratios and allocated instructional time for multihandicapped students. Exceptional Children, 5, 289-296.

Thurlow, M. L., Ysseldyke, J. E., Wotruba, J. W., & Algozzine, B. (1993). Instruction in special education classrooms under varying student-teacher ratios. The Elementary School Journal, 93, 305-321.

Torgeson, J. K., Wagner, R. K., Rashotte, C. A., Alexander, A. W., & Conway, T. (1997). Preventive and remedial interventions for children with severe reading disabilities. Learning Disabilities: A Multidisciplinary Journal, 8(1), 51-61.

Vaughn, S., Moody, S. W., & Schumm, J. S. (1998). Broken promises: Reading instruction in the resource room. Exceptional Children, 64, 211-225.

Westling, D. L., Ferrell, K., & Swenson, K. (1982). Intraclassroom comparison of two arrangements for teaching profoundly mentally retarded children. American Journal of Mental Deficiency, 86, 601-608.

Wexler, E., Izu, J., Carlos, L., Fuller, B., Hayward, G., & Kirst, M. (1998, March). California's class size reduction: Implications for equity, practice, and implementation. [On-line]. Available: www.wested. org/policy/pubs/full_text/class_size/main.htm [1998, Nov. 12].

Wheeler, J. (1993). Kansas class size/caseload study [On-line]. KS: Northeast Kansas Education Service Center. Available: http://www.nekesc.k12ks. us/cscl/fullreport.html [1999, Jan. 22].

Wisniewski, L., & Gargiulo, R. M. (1997). Occupational stress and burnout among special educators: A review of the literature. Journal of Special Education, 31, 325-347.

Ysseldyke, J. E. (1988). Student-teacher ratios and their relationship to instruction and achievement for mildly handicapped students (Final Project Report, Monograph No. 9, Department of Education Grant No. G008630121). Minneapolis: University of Minnesota. (ERIC Document Reproduction Service No. ED 306 720)

Ysseldyke, J. E., & Algozzine, B. (1995). Special education: A practical approach for teachers (3rd ed.). Boston, MA: Houghton-Mifflin.(*)

Books Now

(*) To order books referenced in this journal, please call 24 hrs/365 days: 1-800-BOOKS-NOW (266-5766) or 1-732-728-1040; or visit them on the Web at Children.htm. Use Visa, M/C, AMEX, Discover, or send check or money order + $4.95 S&H ($2.50 each add'l item) to: Clicksmart, 400 Morris Avenue, Long Branch, NJ 07740; 1-732-728-1040 or FAX-1-732-728-7080.

Correspondence concerning this article should be sent to Bert Chiang, Professor of Special Education, University of Wisconsin-Oshkosh, College of Education and Human Services, 800 Algoma Boulevard, Oshkosh, WI 65901. (E-mail:

The development of this monograph was supported in part by a grant from the Wisconsin Department of Public Instruction.

Manuscript received August 1999; accepted April 2000.

SUZANNE RUSS, Research Associate, BERTTRAM CHIANG (CEC #31), Professor of Special Education; BILLIE JO RYLANCE, (CEC #31), Assistant Professor of Special Education; and JOYCE BONGERS, Research Associate, University of Wisconsin, Oshkosh.
COPYRIGHT 2001 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:Exceptional Children
Date:Jan 1, 2001
Previous Article:Asperger Syndrome: The Emerging Challenge to Special Education.
Next Article:Rewarded by Punishment: Reflections on the Disuse of Positive Reinforcement in Schools.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters