EEC550

95R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006, Pages 95–104

Comparing Characteristics of High-Incidence Disability Groups

A Descriptive Review

E D W A R D J . S A B O R N I E , C H A N E V A N S , A N D D O U G L A S C U L L I N A N

A B S T R A C T

More than 2 decades ago, Hallahan and Kauffman and others suggested a cross-categorical approach to teaching students identified with high-incidence disabilities (i.e., emotional– behavioral disabilities, learning disabilities, and mild intellectual dis- abilities) because their behavioral and academic characteristics were seen to be more similar than different. Since that time, more than 150 articles and other works have discussed and compared the characteristics of students across these high-incidence disabil- ity categories. This descriptive review examined 34 studies com- paring various characteristics of students with high-incidence disabilities. The results indicate that students with emotional– behavioral disabilities, learning disabilities, and mild intellectual disabilities do not differ markedly in social adjustment, but do dis- play considerably different cognitive and behavioral profiles. Implications for instruction and placement are discussed.

THE SIMILARITIES AND DIFFERENCES OF STU- dents in the three categories of high-incidence disabilities— emotional–behavioral disorders (EBD), learning disabilities (LD), and mild intellectual disabilities (MID)—have been of interest to teachers, psychologists, researchers, and school professionals for some time. Part of the interest in high- incidence disabilities lies in the fact that more than 70% of all youth with disabilities are served in these three groups. Over the years, research has shown that students with high- incidence disabilities share characteristics in areas such as

social skills (Gresham, Elliot, & Black, 1987), time on task (O’Sullivan, Ysseldyke, Christenson, & Thurlow, 1990), aca- demic growth (Gajar, 1980; Margalit & Roth, 1989), and interactions with peers (Moore & Simpson, 1983). When as- signed to cross-categorical special education classrooms or general education classes, they also share the same teachers, academic curricula, and behavioral programs.

More than one quarter of a century ago, Hallahan and Kauffman (1976, 1977) proposed using a cross-categorical or noncategorical approach to teaching students with mild or high-incidence disabilities, grouping them according to their behavioral characteristics rather than by their disability label. In their seminal article “Labels, Categories, Behaviors: ED, LD, and EMR Reconsidered” (Hallahan & Kauffman, 1977), they asserted that students identified in any one of the three high-incidence categories were more alike than different in their needs for academic and behavioral instruction. Among other factors, Hallahan and Kauffman pointed to confusing definitions in each of the three categories as one justification for grouping them together. The standards for identification in the late 1970s seemed to sometimes depend on who was doing the assessment, which instruments were used, and whether a student did not manifest a specific trait. In com- paring student behavior, Hallahan and Kauffman (1977) con- sidered several characteristics: IQ, underachievement, and personality and social adjustment. By examining the research of the time (e.g., Balthazar & Stevens, 1975; Zigler, 1975),

96 R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006

three groups of high-incidence disabilities cannot be attained. Our goal in this review was to find an answer to the follow- ing modest question: Are students in the three categories of high-incidence disabilities more alike than different?

METHOD

Participants

Students and youth who participated in the studies reviewed herein had been identified as having EBD, LD, or MID by traditional means (i.e., individual state, federal, or similar definitions). The academic settings included general educa- tion, resource rooms, self-contained classrooms, and separate schools. Students’ grade levels ranged from kindergarten to 12th grade, and ethnicities included African American, European American, Hebrew, and Latino. In the 22 studies that specified gender, more than 65% of the participants were boys, and approximately 50% of the participants were elementary-age children. Of the total of 34 studies found for the present review, 13 studies examined issues pertaining to students with EBD, LD, and MID; 7 studies considered aspects covering only groups with LD and MID; and 14 empirical reports provided data concerning groups with EBD and LD only. No studies considered a comparison between groups with EBD and groups with MID on any variable of interest.

Most studies used for this examination included as par- ticipants actual students or youth with EBD, LD, or MID (e.g., Clarizio & Bernard, 1981; Slate, 1995). A few inves- tigations (e.g., Algozzine, Morsink, & Algozzine, 1988; Boucher & Deno, 1979), however, included teachers, parents, or other adults as participants who were asked to categorize or rate various traits found in the children with EBD, LD, and MID.

Literature Search

The Educational Resources Information Center (ERIC), EdInfo, Education Abstracts, and PsycInfo electronic data- bases were used to find relevant studies from 1977 to the present in which students in the categories of high-incidence disabilities were the participants or foci of the study. A va- riety of keywords and descriptors were used to obtain the correct participants and relevant studies, such as emotional disturbance, psychiatric disturbance, educable mental retar- dation, and so on. A manual search of the reference lists from the obtained studies was also conducted to ensure that the related literature exploration was as comprehensive as possi- ble. Our search for related literature produced a total of 152 published reports, including peer-reviewed journal articles, chapters, dissertation abstracts, and monographs. Of these, 58 empirical studies found in peer-reviewed journals were used

they concluded that instructional grouping based on disabil- ity labels was “educationally useless” (p. 146) and recom- mended a noncategorical model driven by students’ specific academic, behavioral, and social deficits.

The interest in high-incidence disability identification, characteristics, placement, and programming has flourished since the 1970s (see also Reschly, Tilly, & Grimes, 1999). Since 1977, more than 150 studies and other works examin- ing similarities and differences across students with EBD, LD, and MID have been published (Sabornie, Cullinan, Osborne, & Brock, 2005). Sabornie et al. conducted a meta- analysis to determine if quantitative research could meaning- fully separate students with EBD, LD, and MID across the domains of IQ, academic achievement, and behavior. They were also interested in determining whether Hallahan and Kauffman’s (1977) conclusions concerning cross-categorical similarities were still true today. Sabornie et al. found 58 group research studies that included the necessary data to compare characteristics of students, all of whom had been identified by state and federal criteria, in high-incidence dis- ability categories. The participants ranged from preschool to 12th grade, and school placements included general educa- tion classrooms, resource rooms, and special education self- contained settings. The meta-analysis examined the three disability groups with a pairwise assessment of LD versus MID, MID versus EBD, and LD versus EBD comparisons. Sabornie et al. found an effect size (ES) of 2.36 for compar- isons involving students with LD versus those with MID in IQ, a relatively robust effect size of −2.17 for comparisons of students with MID versus EBD in IQ, and a very weak effect size of 0.14 for the LD versus EBD comparison in IQ. Regarding IQ, then, students with MID were not at all simi- lar to those with EBD or LD. Academic achievement ESs were significantly different across all comparisons involving the three groups (favoring students with LD and those with EBD over students with MID), and the ESs involving the behavior of students with EBD were found to significantly favor both students with LD and those with MID. Although the meta-analysis results of Sabornie et al. are not the main focus of this article, they will be included in the discussions that follow for comparison purposes.

The purposes of this article are (a) to examine additional cross-categorical studies not found in other reports and (b) to further illuminate the characteristics of students in the three high-incidence disability groups. In other words, the intent is to closely examine the published research concerning stu- dents in the high-incidence categories in the same way in which Hallahan and Kauffman (1976, 1977) did so originally—by using a descriptive review of findings. The studies not found in Sabornie et al. (2005) and reviewed here provide substantive and unique information concerning stu- dents identified as having EBD, LD, and MID. Without a comprehensive review of all the available research literature, using both meta-analytical and traditional review techniques, a complete picture of the similarity or dissimilarity across the

97R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006

in the Sabornie et al. (2005) meta-analysis; 34 additional studies from the same sources could not be included in this meta-analysis and, thus, were included in the present, descriptive review. The following a priori selection criteria were used for studies to be included in the present examina- tion:

1. The study included a comparison of at least two of the three groups of high-incidence dis- ability (i.e., EBD vs. LD, EBD vs. MID, LD vs. MID) on a relevant dependent variable.

2. The study was a group design investigation, with group contrast statistics reported.

3. Each study had to appear in a peer-reviewed journal. In an attempt to control for study quality—an elusive concept with little agree- ment among experts (see Wortman, 1994)—we eschewed research literature (e.g., disserta- tions) that was not found in peer-reviewed journals. The 34 studies selected in the present review of the literature were found in 21 differ- ent peer-reviewed journals.

The Sabornie et al. (2005) meta-analysis selected three domains of interest for dependent variable comparisons across the categories of high-incidence disabilities: IQ, aca- demic achievement, and behavior. In the present descriptive review, we were able to categorize the available studies into the following realms: IQ, academic, social, behavior, func- tional skills, and other. The other category encompassed four subtypes of investigations: student characteristics (not associ- ated with the other five domains), parent perceptions, teacher attitudes and decisions, and ethnicity and socioeconomic sta- tus (see Table 1). Studies concerning postschool status were included in the functional skills domain. Data from one of the 34 studies (i.e., Margalit, 1989) were used in more than one domain of interest (i.e., academic and behavior). The first author and two doctoral students in special education were involved in the literature searches and in coding studies into the available domains of interest. Discussions between the three primary literature searchers led to concurrence when there was disagreement regarding the placing of a study into a suitable domain of interest for this review.

RESULTS Table 1 presents a summary of the studies used in the present descriptive review.

Intelligence Comparisons

According to federal regulations, to be eligible for the in- tellectual disability category, a student must exhibit signifi- cantly below-average general intellectual functioning along

with deficits in adaptive behavior. Although individual states interpret this requirement differently, a nationwide survey (Denning, Chamberlain, & Polloway, 2000) revealed that 60% of the states used an intelligence test cutoff score of 70, or a score of at least 2 standard deviations below the mean, to meet the intellectual functioning criterion. In order for a stu- dent to qualify as having either EBD or LD, however, aca- demic shortfalls must not be caused primarily by intellectual deficits or mental retardation.

Hallahan and Kauffman (1976, 1977) stated that chil- dren with EBD and LD also had below-average IQs. In the mid-1970s, they surveyed more than 100 students with LD and found the mean IQ to be 91 (Hallahan & Kauffman, 1977). Research studies from the 1970s to the present have continued to show that students with EBD have IQs in the low-average range, with a mean of about 90 to 95 (Kauffman, 2005).

Three of the research comparisons in this review (Clar- izio & Bernard, 1981; Petersen & Hart, 1978; Slate, 1995) in- cluded cognitive ability scores of students in the high-incidence disability groups. As a group, these studies found that stu- dents with MID had substantially lower cognitive ability scores than students with EBD and LD. Slate also found that students with MID and LD differed in the degree of dis- crepancy between Wechsler Intelligence Scale for Children– Revised (WISC-R) Verbal and Performance IQ scores, with the group with LD demonstrating larger Verbal–Performance IQ discrepancies.

In the Sabornie et al. (2005) meta-analysis, IQ findings from 25 studies compared mean IQ scores of students with MID and LD; 11 studies included IQ comparisons of students with EBD and those with MID; and 26 investigations com- pared IQ functioning across EBD and LD groups. The mean ES estimates for IQ with both LD versus MID and MID ver- sus EBD comparisons were found to be quite substantial (2.36 and −2.17, respectively; the negative ES favors EBD over MID). In contrast, the IQ comparison between LD and EBD was very weak, with an average weighted ES estimate of only 0.14.

Academic Domain Comparisons

Lack of achievement in school is typically one of the first indicators of a cognitive disorder (Beirne-Smith, Ittenbach, & Patton, 2006) and one of the five characteristics of emotional disturbance. It is no surprise, therefore, to find students with EBD, LD, and MID with low academic performance. Halla- han and Kauffman (1976, 1977) maintained that within this construct, there was more overlap than uniqueness among students in the three groups.

The majority of the articles reviewed here concurred with the original conclusions of Hallahan and Kauffman (1976, 1977) regarding academic achievement. Seven of the 11 comparisons indicated no significant differences among students with high-incidence disabilities on academically

98 R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006

TABLE 1. High-Incidence Disability Comparison Studies Reviewed

Study Domain Variable Results

Clarizio & Bernard (1981) IQ WISC-R subtest scores EBD = LD LD > MID

Petersen & Hart (1978) IQ WISC-R LD, EBD > MIDa

LD = EBD

Slate (1995) IQ WISC-III, WISC-R LD > MID

Jaben (1986) Academic Creative thinking LD = EBD

Margalit (1989) Academic Academic competence LD = EBD

Margalit & Roth (1989) Academic Spelling LD = MID

Ysseldyke et al. (1991) Academic Time on academics MID < LD, EBD LD = EBD

Downing et al. (1990) Social Mainstream social skills LD > EBD on some skills LD = EBD on other skills

Margalit (1995) Social Loneliness LD = EBD

Moore & Simpson (1983) Social Verbal statements LD = EBD

Elmquist et al. (1992) Behavior Alcohol use LD < EBD

Gadow et al. (1990) Behavior Aggression LD < EBD

Handwerk & Marshall (1998) Behavior Problem behaviors LD < EBD and LD = EBD on inattention

Margalit (1989) Behavior Social adjustment LD > EBD and LD < EBD on social maladjustment

McConaughy et al. (1994) Behavior Behavior problems LD < EBD

Sindelar et al. (1985) Behavior Deviant behaviors LD < EBD

Affleck et al. (1990) Functional skills Postschool status MID, LD > EBD

Botuck et al. (1996) Functional skills Job placement LD = MID

Levy et al. (1993) Functional skills Job training LD = MID in employment LD < MID in wages

Rimmerman et al. (1995) Functional skills Employment problems LD = MID

Algozzine et al. (1998) Other Teachers’ instructional practices LD, EBD > MID

Barona et al. (1993) Other SES and ethnicity MID > LD, EBD

Bernard & Clarizio (1981) Other SES LD = EBD = MID

Boucher & Deno (1979) Other Teachers’ perceptions of characteristics LD = EBD

Teachers’ goals LD = MID

Elliott & Boeve (1987) Other Ethnicity LD = EBD = MID

Gelb & Mizokawa (1986) Other SES LD, MID > EBD

Leigh et al. (1995) Other Teachers’ instructional issues LD < EBD, MID

MacMillan et al. (1998) Other Ethnicity LD > MID

Morgan (1986) Other Locus of control LD > EBD internal locus of control

Potter et al. (1983) Other Identification LD > EBD, MID

Rizzo & Kirkendall (1995) Other Attitudes of preservice teachers Older persons less tolerant of EBD

Simpson & Myles (1989) Other Mainstream modifications LD = MID = EBD

Singh et al. (1994) Other Medication treatment LD = EBD

Stern et al. (1991) Other Drug use among family members LD < EBD

Ysseldyke & Algozzine (1981) Other Referral problems LD, MID < EBD

Note. EBD = emotional–behavioral disorders; LD = learning disabilities; MID = mild intellectual disabilities; SES = socioeconomic status; WISC-R = Wechsler Intel- ligence Scale for Children–Revised; WISC-III = Wechsler Intelligence Scale for Children, third edition. a Interpretation: In this study from the IQ domain, students identified as EBD and LD were both shown to score higher than those with MID on the WISC-R, and stu- dents with LD and EBD were found to be statistically similar in scores on the WISC-R.

99R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006

oriented variables. Four studies compared students with LD with those with MID on academic acquisition or improve- ment after having been taught a specific intervention. Although individuals with MID made slightly fewer gains than those with LD, there was no significant difference in stu- dents’ posttest scores in skills such as spelling performance (Margalit & Roth, 1989).

Two additional studies compared elementary-age stu- dents with EBD and those with LD in academic competence (Margalit, 1989) and creative thinking skills (Jaben, 1986). Here again, no significant differences were found between the two disability groups. Jaben used a treatment-control model to teach divergent thinking skills to 98 students in self- contained EBD and LD classrooms. The treatment group, consisting of students with both EBD and LD, received 14 weeks training in creative processing techniques, while the control group was offered no additional intervention. Jaben used the Torrance Tests of Creative Thinking as a pretest– posttest measure of verbal fluency, flexibility, and originality. Students with EBD and LD in the treatment group scored sig- nificantly higher than did students with the same disabilities who received no intervention, with no substantial difference between the students with EBD and those with LD within each group.

Ysseldyke, Thurlow, Christenson, and Muyskens (1991) explored classroom and home environments using qualitative and quantitative methods to compare the levels of academic activity and emphasis for 122 students with EBD, LD, and MID. It was found repeatedly in different settings (separate classroom, resource room) and for different content areas (math, reading, and writing) that students with MID received significantly less academic instruction and were given signif- icantly more free-time activities than students with EBD or LD. When home environments were examined, it was found that families of students with EBD were rated lower on secu- rity and valuing education than were families of students with LD. Supporting school effort was also rated lower for fami- lies of students with EBD than for those of students with LD or MID.

Sabornie et al. (2005) examined findings from 33 stud- ies to determine if an ES difference among groups with EBD, LD, and MID could be discerned within the academic achievement domain. Unlike IQ, the mean effect sizes indi- cated only moderate differences for one of the comparisons, LD versus MID (ES = 0.42, favoring the group with LD), but a strong difference between students with MID and those with EBD (ES = −0.70, favoring those with EBD). There was a weak but statistically significant difference between stu- dents with EBD and LD, with an ES of −0.22 favoring stu- dents with LD.

Social Domain Comparisons

Hallahan and Kauffman (1976) concluded that even though social adjustment deficits were generally associated with EBD,

students with LD and MID were also lacking in this area. Referring to the American Association on Mental Disability’s (AAMD’s) criterion for adaptive behavior deficiency, they stated that it was “logical to assume that personality and social adjustment are synonymous with adaptive behavior” (p. 144). Therefore, they concluded that students with MID would share this characteristic with those with EBD and LD.

The results from the three studies comparing 10 vari- ables for students with EBD and those with LD only in the social domain were equivocal, with one study (Downing, Simpson, & Myles, 1990) showing significant differences, but two additional investigations (Margalit, 1995; Moore & Simpson, 1983) indicating no differences between the two groups. Downing et al. (1990) asked 56 general and special education teachers to rate students with EBD and with LD on their proficiency and use of nonacademic skills, including asking for help, avoiding swearing and fighting, obeying rules, and interacting with teachers. Students with LD were rated higher on following school and classroom rules, but equal to those with EBD in cursing, fighting, and interacting with teachers. In contrast, no significant differences were found in verbal interactions among 45 students with EBD or LD and their peers and teachers (Moore & Simpson, 1983). Students with EBD and LD both emitted significantly more neutral and negative than positive comments to peers. They also made more negative comments to their teachers than did their nondisabled peers.

Behavior Domain Comparisons

As with social skill deficits, it is expected that students with a primary identification of EBD will exhibit more problem- atic behaviors than students with LD or MID. On a behavioral continuum, maladaptive episodes of students with EBD are experienced more often and are more intense and longer last- ing than are those of students whose primary diagnosis is either LD or MID. Although Hallahan and Kauffman (1976) agreed that “subtle differences do arise in the frequency with which certain behaviors are exhibited” (p. 36), they main- tained that students with LD and MID also exhibit behavioral deficits in areas of interpersonal relationships and social– emotional adjustment.

Our review of six studies comparing the behaviors of students in high-incidence disability groups found that stu- dents with EBD exhibited more acting out, hyperactivity, social maladjustment (Margalit, 1989), externalizing and in- ternalizing behaviors (Handwerk & Marshall, 1998), and illi- cit drug and alcohol use (Elmquist, Morgan, & Bolds, 1992) than did students identified with LD or MID. The vast major- ity of the behavior variable comparisons concurred that stu- dents with EBD had significantly more behavioral problems than students with LD or MID. Margalit (1989) used teacher rating scales to measure the adjustment of 73 boys with either EBD or LD. She found that boys identified as having EBD were more hyperactive and socially maladjusted than were

100 R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006

boys identified with LD. Although not as hostile or aggres- sive, the boys with LD were still seen as having interpersonal relationship deficits. Other studies concurred that although students with LD experienced behavioral problems, students identified with EBD had significantly more problems with aggression, self-destructive behaviors, withdrawal, and inat- tention (Handwerk & Marshall, 1998; McConaughy, Matti- son, & Peterson, 1994; Sindelar, King, Cartland, Wilson, & Meisel, 1985).

In contrast with these results, Gadow, Sprafkin, and Grayson (1990) found in their validation study of the Help– Hurt Game that 38 elementary-age students with EBD and LD exhibited no significant differences. The students, ages 5 to 10, were randomly divided into two groups, each contain- ing students with both disabilities. One group was shown a cartoon with a high content of aggressive action, while the other watched one filled with altruistic deeds. Then, each child played a game where he or she could either “help” or “hurt” a fictitious playmate. The determining variable for aggression seemed to be the cartoon program—either violent or helpful—and not membership in a disability group.

Elmquist et al. (1992), included here in the behavior domain, examined alcohol and drug use among adolescents with high-incidence disabilities. Elmquist et al. used a self- report measure with 110 sixth to twelfth graders with EBD and LD to determine the extent of alcohol use. The results of the survey showed no significant difference in use among stu- dents with EBD versus students with LD who were taught in resource or general education classes. Students with EBD who were taught in self-contained settings, however, reported a significantly higher use of both alcohol and illegal drugs than either students with LD or students with EBD served in less restrictive settings.

Sabornie et al. (2005) considered 41 comparisons for evidence of similarity or difference in behavior among stu- dents with EBD, LD, and MID. They found the ES estimate to be nugatory (−0.01) for the behavioral domain comparison of students with LD versus those with MID. In the MID ver- sus EBD comparison (ES = 0.70), students identified as hav- ing EBD exhibited significantly greater behavioral problems. The third comparison—LD versus EBD—showed another strong and statistically significant difference between the groups, with an ES estimate of 0.72, favoring better behavior among those with LD. These findings correspond to those from the present descriptive review in that students with EBD had significantly more behavior problems than students in the other categories.

Functional Skill Comparisons

The construct of functional skills is particularly germane to students with MID, and just as synonymous with adaptive skills as “personality and social adjustment” (Hallahan & Kauffman, 1977, p. 144). The American Association on Men-

tal Retardation (AAMR) listed 10 domains of adaptive skills in its 1992 definition of intellectual disability (Luckasson et al., 1992), and functional utility is implicit throughout the descriptions of self-direction, health and safety, leisure, and work. The present review examined four studies concerning the acquisition of functional skills or postschool status across the three high-incidence disability groups.

Although dismal postschool outcomes for students with EBD are widely recognized (e.g., Greenbaum et al., 1996; Oswald & Coutinho, 1996; Wagner, 1995), none of these studies compared students with EBD to those with LD or MID. We found three studies (Botuck, Levy, & Rimmerman, 1996; Levy et al., 1993; Rimmerman et al., 1995) that were associated with results from an adult employment training program, the Young Adult Institute. This project offered indi- viduals preplacement assessment and training before begin- ning work. There was no indication that the disabilities of individuals with LD or MID were important factors in ini- tial job placement or types of problems experienced on the job. Differences between salaries were noted, however, with young adults with LD receiving higher wages than those with MID.

In a postschool status investigation (Affleck, Edgar, Le- vine, & Kortering, 1990), former students with LD and those with MID were contacted twice—at 6 months and 30 months—after either graduating or “aging out” of special education programs in a statewide telephone survey. Ques- tions were asked concerning postsecondary education, employment, salary, and independent living. Individuals with LD rated their status on all variables significantly higher than did young adults with MID.

Other Comparisons

Research in the last domain includes results from studies (n = 15) concerning the perceptions of students, parents, and educators on such diverse topics as locus of control (Morgan, 1986), parent requests for general education modifications (Simpson & Myles, 1989), presence of alcohol abuse among parents (Stern, Kendall, & Eberhard, 1991), and knowledge of psychotropic medications (Singh, Epstein, Stout, Luebke, & Ellis, 1994). Also integrated in this catch-all category are opinions about the characteristics of EBD, LD, and MID by preservice teachers (Rizzo & Kirkendall, 1995), school psy- chologists and social workers (Ysseldyke & Algozzine, 1981), and general and special educators (Boucher & Deno, 1979), and research addressing bias based on socioeconomic status (SES) and ethnicity.

In a comparison of locus of control (Morgan, 1986), stu- dents with LD gave themselves more credit for success in areas of intellectual, social, and physical abilities than did students with EBD. Students with EBD attributed their suc- cess to either fate or chance and showed a substantially exter- nal locus of control.

101R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006

When parents of students with EBD, LD, and MID were questioned about the number of services and modifications they wanted in planning for the success of their children in inclusion classes (Simpson & Myles, 1989), no significant difference was found among the groups. Moreover, in another study (Algozzine, Morsink, & Algozzine, 1998), significant difference was not shown in most of the instructional strate- gies and techniques used by 40 self-contained classroom teachers in classrooms with students with EBD, LD, and MID. Only one discrepancy was observed among the three sets of teachers: Those who taught students with MID modi- fied their instruction about half as frequently as teachers of students with EBD or LD.

Six studies considered questions of bias in the identifi- cation of a high-incidence disability based on race or socio- economic factors. Three of these studies found that being classified as having MID or LD was positively related with ethnicity or SES (Barona, Santos de Barona, & Faykus, 1993; Gelb & Mizokawa, 1986; MacMillan, Gresham, & Bocian, 1998). Gelb and Mizokawa, for example, in examining rec- ords of school-age students in the United States, found that MID were correlated with low SES and students of African American and Latino descent, whereas LD were associated with high SES and a European American heritage. Neither connection was found for students with EBD, however. The remaining studies (Bernard & Clarizio, 1981; Elliott & Boeve, 1987) found no ethnic or SES bias in identification practices involving high-incidence disabilities.

Stern et al. (1991) explored the hypothesis that students with EBD were more likely to have parents with alcohol addiction than students with LD or without a disability. To test this theory, 30 first- to fifth-grade students in each of 3 groups—EBD, LD, and no disabilities—were randomly selected to take the Children of Alcoholics Screening Test, a measure of their attitudes and perceptions about their parents’ drinking habits. Statistically significant differences were not found among the students’ ratings; however, the disparity among groups was noteworthy. An immediate family member with chemical dependency was reported by 47% of students with EBD, by 15% of those with LD, and by 20% of students without disabilities.

DISCUSSION Hallahan and Kauffman (1976, 1977) asserted that the dis- similarity in the characteristics of students with high- incidence disabilities was negligible, even though they noted deficits of slightly differing degrees in IQ, academic achieve- ment, and personality and social adjustment across the three groups. They called for a discontinuation of homogeneous grouping by administrative category and proposed instead to implement a task-centered model that would remediate stu- dents’ specific academic and behavioral deficits, no matter

what their special education label. Our intent in this review was not to debunk or criticize the conclusions of Hallahan and Kauffman from the late 1970s. Nevertheless, some of the results we found differ from those of Hallahan and Kauffman.

Limitations Before summarizing the results of our review, a few limita- tions of our methodology are worthy of discussion. We ini- tially eliminated single-subject studies and reports without recorded statistical significance in our attempt to draw mean- ingful conclusions and implications from this review. How- ever, knowing that statistical significance is partly a function of sample size, we proceeded with caution. Kavale (2001) warned of the vagaries associated with using a “box score” approach in the type of analysis used herein. Simply record- ing a yes or a no for the statistical significance of each study, tallying the totals, and proclaiming a “winner” does not suf- ficiently address the subtleties of individual results. We found this to be true in the domain of behavior. Students with EBD were found unequivocally to exhibit more maladaptive behaviors than did students with LD, but on closer examina- tion, the studies confirmed that students with LD also had difficulty with social relationships, hyperactivity, and inatten- tion.

An additional shortcoming of our review concerns the operational parameters of a descriptive review compared to a meta-analysis. A true strength of the meta-analytic technique is its ability to assist in removing bias in the interpretation of findings (Kavale, 2001; Lipsey & Wilson, 2001; Wang & Bushman, 1999). This trait of meta-analysis, unfortunately, is lost in a descriptive review of the research literature. By combining our descriptive review with the available meta- analysis, however, we attempted to provide a comprehensive addition to the extant discourse concerning students with high-incidence disabilities.

Group Differences and Similarities Despite its limitations, our present compilation helps com- plete the review of research since the late 1970s. The com- parison of students with EBD and those with MID showed that 80% of the studies we included here indicated a signifi- cant difference in IQ, academic skills, and behavior prob- lems. The meta-analysis of Sabornie et al. (2005) also dem- onstrated a strong disparity between these two groups in IQ (ES = −2.20), academic achievement (ES = −0.70), and behavior problems (ES = 0.73). In other words, students with EBD, when compared to those with MID, have higher IQs, superior academic skills, and more behavioral problems. The research seems to support that when students with EBD and MID are compared side by side, they are not so similar on these important constructs for special education identification and instruction. Key questions related to these differences are whether the disparity is worthy of educating students in these

102 R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006

two groups apart from each other, and whether such separa- tion is cost effective and instructionally efficacious.

Sabornie et al. (2005) found very strong to moderate ESs between students with MID and those with LD in the domains of IQ (ES = 2.36) and academic achievement (ES = 0.42). These ES estimates favored students with LD in each domain, meaning that such students were more skilled than their peers with MID. Within our descriptive review, as expected, the greatest differentiation between these groups was in the IQ domain. Examination across the other domains showed that 67% of the studies indicated significant differ- ences between students with LD and MID. These data were examined further to discern nuances in differences and simi- larities. In the academic domain, 50% of the studies showed no difference (in spelling and story recall); one study favored students with MID (in reading), and one study favored stu- dents with LD (in time spent on academic tasks). Therefore, in our contrast between students with LD and MID, with the exception of IQ, the differences appear to be far from remark- able.

Our results also suggest specific problem areas for stu- dents in each of the disability groups. In addition to IQ deficits, students with MID made fewer gains with academic tasks, were given less teacher attention, and rated themselves as having low postschool status. Strengths were noted when students with MID were compared to those with LD (in read- ing) and EBD (in social skill and behavior). Thus, in a rela- tive sense, students with MID have positive qualities. This conclusion fits with the recent emphasis and movement toward strength-based assessment for students with high- incidence disabilities (see Epstein & Sharma, 1998; Reid, Epstein, Pastor, & Ryser, 2000).

In comparing the strengths and needs of students with EBD and LD, we found that although their IQ scores were comparable, students with LD had greater deficits in reading comprehension and general academic competency. Similari- ties were seen in verbal skills, creative thinking, and time spent on academics. Behavior problems and lack of social skills were demonstrated by both groups of students, al- though youngsters with EBD experienced significantly more problems in these domains.

One question emerges when examining our results: How many studies have to show verifiable differences among the present comparison groups before one can say that the differ- ences are convincing? What is needed is a descriptive review interpretation rubric similar to what has been developed for meta-analysis (see Cohen, 1988; Forness, Kavale, Blum, & Lloyd, 1997). Although the cutoff points are somewhat arbi- trary, Forness et al. and Cohen suggested that ES estimates of a certain magnitude indicate weak, moderate, and strong dif- ferences. Given that the published research literature favors the reporting of statistically significant differences between comparison groups, a strength rubric would assist those con- ducting descriptive reviews of the literature. If a certain per- centage of studies showing significant differences between

groups could be similarly labeled with magnitude, this rubric could reduce some of the interpretation bias in descriptive reviews of the literature such as the present one.

Practical and Educational Implications

The results of this descriptive review lead to another impor- tant question: Are these findings educationally relevant? One could argue that the most basic information that teachers need to have is the who found in special education—one of the per- petual issues facing the field (Bateman, 1994; Kauffman & Landrum, 2006). Teachers must know what to expect con- cerning students in different categories of exceptionality, so that these students’ needs can be met. We believe that the present results support differentiated curricular and instruc- tional approaches by category of high-incidence disability, but not at the level of any one student. Instead, the present results may have more implications for teacher preparation and university program planning. It appears that preservice and inservice teachers need to know that students with EBD need more behavioral support than do their peers with LD or MID, and students with MID need more assistance concern- ing the cognitive and academic domains in comparison to stu- dents with EBD or LD. Teachers in training should be made aware of the categorical differences shown herein long before accepting employment in schools.

When we examine the results of our review using the original framework of Hallahan and Kauffman in the 1970s, we conclude that students with EBD, LD, and MID do not differ patently in social adjustment, but do display substan- tially different IQ and behavioral characteristics. That IQ separated the comparison groups is no surprise, given the def- inition of intellectual disability. Students with EBD and MID seemed to be the most different when all the comparisons and variables of interest are considered in totality. The recom- mendation of Hallahan and Kauffman for task-centered in- struction based on behavioral characteristics continues to be an appropriate view for educating students. Individualized, appropriate assessment and effective instruction regardless of the disability category are still best practice for students with high-incidence disabilities in special and general educa- tion.�

EDWARD J. SABORNIE, PhD, is professor in the Department of Cur- riculum and Instruction at North Carolina State University, Raleigh, North Carolina. His research interests include similarities and differences across students with high-incidence disabilities, social skills, and secondary-level instructional issues in special education. CHAN EVANS, PhD, is an assis- tant professor of special education at Augusta State University, Augusta, Georgia. Her research interests include characteristics of and interventions for young students with emotional–behavioral disorders. DOUGLAS CUL- LINAN, EdD, is professor in the Department of Curriculum and Instruction at North Carolina State University. His research interests include the nature of and intervention for emotional and behavior disorders of students. Address: Edward J. Sabornie, North Carolina State University, College of Education, Campus Box 7801, Poe Hall, Suite 602, Raleigh, NC 27695- 7801.

103R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006

REFERENCES

*Study included in the descriptive review.

*Affleck, J. Q., Edgar, E., Levine, P., & Kortering, L. (1990). Postschool sta- tus of students classified as mildly mentally retarded, learning disabled, or nonhandicapped: Does it get better with time? Education and Train- ing of the Mentally Retarded, 25, 315–324.

*Algozzine, B., Morsink, C. V., & Algozzine, K. M. (1998). What’s happen- ing in self-contained special education classrooms? Exceptional Chil- dren, 55, 259–265.

Balthazar, E. E., & Stevens, H. A. (1975). The emotionally disturbed, men- tally retarded. Englewood Cliffs, NJ: Prentice Hall.

*Barona, A., Santos de Barona, M., & Faykus, S. P. (1993). The simultane- ous effects of sociocultural variables and WISC-R factors on MR, LD, and nonplacement of ethnic minorities of special education. Education and Training of the Mentally Retarded, 28, 66–74.

Bateman, B. D. (1994). Who, how, and where: Special education’s issues in perpetuity. The Journal of Special Education, 27, 509–520.

Beirne-Smith, M., Ittenbach, R. F., & Patton, J. R. (2006). Mental retarda- tion (7th ed.). Upper Saddle River, NJ: Merrill/Prentice Hall.

*Bernard, R., & Clarizio, H. (1981). Socioeconomic bias in special educa- tion placement decisions. Psychology in the Schools, 18, 178–183.

*Botuck, S., Levy, J. M., & Rimmerman, A. (1996). Gender-related differ- ences in placements rates of young adults with mental retardation and severe learning disabilities. International Journal of Rehabilitation Research. Internationale Zeitschrift fur Rehabilitationsforschung. Revue Internationale de Recherches de Readaptation, 19, 259–263.

*Boucher, C. R., & Deno, S. L. (1979). Learning disabled and emotionally disturbed: Will the labels affect teacher planning? Psychology in the Schools, 16, 395–402.

*Clarizio, H., & Bernard, R. (1981). Recategorized WISC-R scores of learn- ing disabled children and differential diagnosis. Psychology in the Schools, 18, 5–12.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Mahwah, NJ: Erlbaum.

Denning, C. B., Chamberlain, J. A., & Polloway, E. A. (2000). An evaluation of state guidelines for mental retardation: Focus on definition and clas- sification practices. Education and Training in Mental Retardation and Developmental Disabilities, 35, 226–232.

*Downing, J. A., Simpson, R. L., & Myles, B. S. (1990). Regular and spe- cial educator perceptions of nonacademic skills needed by main- streamed students with behavioral disorders and learning disabilities. Behavioral Disorders, 15, 217–226.

*Elliott, S. N., & Boeve, K. (1987). Stability of WISC-R: An investigation of ethnic differences overtime. Educational and Psychological Measure- ment, 47, 461–465.

*Elmquist, D. L., Morgan, D. P., & Bolds, P. K. (1992). Alcohol and other drug use among adolescents with disabilities. The International Journal of the Addictions, 27, 1475–1483.

Epstein, M. H., & Sharma, J. (1998). Behavioral and emotional ratings scale: A strength-based approach to assessment. Austin, TX: PRO-ED.

Forness, S. R., Kavale, K. A., Blum, I. M., & Lloyd, J. W. (1997). Mega- analysis of meta-analyses: What works in special education and related services. Teaching Exceptional Children, 30(4), 4–9.

*Gadow, K. D., Sprafkin, J., & Grayson, P. (1990). The help–hurt game as a measure of aggression in children with leaning and behavior disorders. Learning and Individual Differences, 2, 337–351.

Gajar, A. H. (1980). Characteristics across exceptional categories: EMR, LD, and ED. The Journal of Special Education, 14, 165-173.

*Gelb, S. A., & Mizokawa, D. T. (1986). Special education and social struc- ture: The commonality of “exceptionality.” American Educational Research Journal, 23, 543–557.

Greenbaum, B. E., Dedrick, R. F., Friedman, R. M., Kutash, K., Brown, E. C., Lardieri, S. P., et al. (1996). National adolescent and child treatment study (NACTS): Outcomes for children with serious emotional and

behavioral disturbance. Journal of Emotional and Behavioral Disorders, 4, 130–146.

Gresham, F. M., Elliot, S. N., & Black, F. L. (1987). Factor structure repli- cation and bias investigation of the teacher rating of social skills. Jour- nal of School Psychology, 17, 82–91.

Hallahan, D. P., & Kauffman, J. M. (1976). Introduction to learning disabil- ities: A psycho-behavioral approach. Englewood Cliffs, NJ: Prentice Hall.

Hallahan, D. P., & Kauffman, J. M. (1977). Labels, categories, behaviors: ED, LD, and EMR reconsidered. The Journal of Special Education, 11, 139–149.

*Handwerk, M. L., & Marshall, R. M. (1998). Behavioral and emotional problems of students with learning disabilities, serious emotional distur- bance, or both conditions. Journal of Learning Disabilities, 31, 327– 338.

*Jaben, T. H. (1986). Impact of instruction on behavior disordered and learn- ing disabled students’ creative behavior. Psychology in the Schools, 23, 401–405.

Kauffman, J. M. (2005). Characteristics of emotional and behavioral disor- ders of children and youth (8th ed.). Upper Saddle River, NJ: Prentice Hall.

Kauffman, J. M., & Landrum, T. J. (2006). Children and youth with emo- tional and behavioral disorders: A history of their education. Austin, TX: PRO-ED.

Kavale, K. A. (2001). Meta-analysis: A primer. Exceptionality, 9, 175–184. *Leigh, J. E., Huntze, S. L., & Lamorey, S. (1995). Topical issues education:

Teaching controversial or sensitive topics to students with learning dis- abilities. Journal of Learning Disabilities, 28, 353–363.

*Levy, J. M., Murphy, B. S., Levy, P. H., Kramer, M. E., Rimmerman, A., & Botuck, S. (1993). Monitoring process and employment outcomes over time: The YAI Employment Project. Journal of Developmental and Physical Disabilities, 5, 167–179.

Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage.

Luckasson, R., Coulter, D. L., Polloway, E. A., Reiss, S., Schalock, R. L., Snell, M. E., et al. (1992). Mental retardation: Definition, classification, and systems of supports. Washington, DC: American Association on Mental Retardation.

*MacMillan, D. L., Gresham, F. M., & Bocian, K. M. (1998). Curing men- tal retardation and causing learning disabilities: Consequences of using various WISC-III tests to estimate aptitude of Hispanic students. Jour- nal of Psychoeducational Assessment, 16, 36–54.

*Margalit, M. (1989). Academic competence and social adjustment of boys with learning disabilities and boys with behavior disorders. Journal of Learning Disabilities, 22, 41–50.

*Margalit, M. (1995). Social skills learning for students with learning dis- abilities and students with behavior disorders. Educational Psychology, 15, 445–456.

*Margalit, M., & Roth, Y. B. (1989). Strategic keyboard training and spelling improvement among children with learning disabilities and mental retar- dation. Educational Psychology, 9, 321–329.

*McConaughy, S. H., Mattison, R. E., & Peterson, R. L. (1994). Behavioral/ emotional problems of children with serious emotional disturbances and learning disabilities. School Psychology Review, 23, 81–98.

*Morgan, S. R. (1986). Locus of control in children labeled learning dis- abled, behaviorally disordered, and learning disabled/behaviorally dis- ordered. Learning Disabilities Research, 2, 10–13.

*Moore, S. R., & Simpson, R. L. (1983). Teacher–pupil and peer verbal interactions of learning disabled, behavior-disordered, and nonhandi- capped students. Learning Disability Quarterly, 6, 273–282.

O’Sullivan, P. J., Ysseldyke, J. E., Christenson, S. L., & Thurlow, M. L. (1990). Mildly handicapped elementary students’ opportunity to learn during reading instruction in mainstream and special education settings. Reading Research Quarterly, 25, 131–146.

Oswald, D. P., & Coutinho, M. J. (1996). Leaving school: The impact of state economic and demographic factors for students with serious emotional

disturbance. Journal of Emotional and Behavioral Disorders, 4, 114– 125.

*Petersen, C. R., & Hart, D. H. (1978). Use of multiple discriminant func- tion analysis in evaluation of a state-wide system for identification of educationally handicapped children. Psychological Reports, 43, 743–755.

*Potter, M. L., Ysseldyke, J. E., Reagan, R. R., & Algozzine, B. (1983). Eligibility and classification decisions in educational settings: Issuing “passports” in a state of confusion. Contemporary Educational Psychol- ogy, 8, 146–157.

Reid, R., Epstein, M. H., Pastor, D. A., & Ryser, G. R. (2000). Strengths- based assessment differences across students with LD and EBD. Reme- dial and Special Education, 21, 346–355.

Reschly, D. J., Tilly, W. D., & Grimes, J. (Eds.). (1999). Special education in transition: Functional assessment and noncategorical programming. Longmont, CO: Sopris West.

*Rimmerman, A., Botuck, S., & Levy, J. M. (1995). Job placement for indi- viduals with psychiatric disabilities in supported employment. Psychi- atric Rehabilitation Journal, 19, 37–43.

*Rizzo, T. L., & Kirkendall, D. R. (1995). Teaching students with mild dis- abilities: What affects attitudes of future physical educators? Adapted Physical Activity Quarterly, 12, 205–216.

Sabornie, E. J., Cullinan, D., Osborne, S. S., & Brock, L. M. (2005). Intel- lectual, academic, and behavioral functioning of students with high- incidence disabilities: A cross-categorical meta-analysis. Exceptional Children, 72, 47-63.

*Simpson, R. L., & Myles, B. (1989). Parents’ mainstreaming modification preferences for children with educable mental handicaps, behavior dis- orders, and learning disabilities. Psychology in the Schools, 26, 292–301.

*Singh, N. N., Epstein, M. H., Stout, C., Luebke, J., & Ellis, C. R. (1994). Psychopharmacological intervention III: A comparison of teacher per- ceptions of psychotropic medication for students with learning disabili- ties or serious emotional disturbance. Journal of Developmental and Physical Disabilities, 6, 101–113.

*Sindelar, P. T., King, M. C., Cartland, D., Wilson, R. J., & Meisel, C. J.

(1985). Deviant behavior in learning disabled and behaviorally disor- dered students as a function of level and placement. Behavioral Disor- ders, 10, 105–112.

*Slate, J. R. (1995). Discrepancies between IQ and index scores for a clini- cal sample of students: Useful diagnostic indicators? Psychology in the Schools, 32, 103–108.

*Stern, R., Kendall, A., & Eberhard, P. (1991). Children of alcoholics in the schools: Where are they? Their representation in special education. Psy- chology in the Schools, 28, 116–123.

Wagner, M. (1995). Outcomes for youths with serious emotional disturbance in secondary school and early adulthood. The Future of Children: Crit- ical Issues for Children and Youths, 5, 90–112.

Wang, M. C., & Bushman, B. J. (1999). Integrating results through meta- analytic review using SAS software. Cary, NC: SAS Institute.

Wechsler, D. (1974). Wechsler Intelligence Scale for Children-Revised. San Antonio, TX: The Psychological Corporation.

Wortman, P. M. (1994). Judging research quality. In H. Cooper & L. V. Hedges (Eds.), The handbook of research synthesis (pp. 97–110). New York: Russell Sage Foundation.

*Ysseldyke, J. E., & Algozzine, B. (1981). Diagnostic classification deci- sions as a function of referral information. The Journal of Special Edu- cation, 15, 429–435.

*Ysseldyke, J. E., Thurlow, M. L., Christenson, S. L., & Muyskens, P. (1991). Classroom and home learning differences between students labeled as educable mentally retarded and their peers. Education and Training of the Mentally Retarded, 26, 3–17.

Zigler, E. (1975). Cognitive–developmental and personality factors in behav- ior. In J. M. Kauffman & J. S. Payne (Eds.), Mental retardation: Intro- duction and personal perspectives. Columbus, OH: Merrill.

Received: 3/29/2002 Initial acceptance: 6/10/2002 Revision received: 8/10/2005 Final acceptance: 8/29/2005

104 R E M E D I A L A N D S P E C I A L E D U C A T I O N Volume 27, Number 2, March/April 2006

CALL FOR MANUSCRIPTS CAREER DEVELOPMENT FOR EXCEPTIONAL INDIVIDUALS Official journal for the Division on Career Development and Transition – CEC

Now published by PRO-ED Articles for CDEI are accepted for review on a continual basis. The editors welcome articles in the areas of secondary education, transition, and career development of persons with disabilities and special needs. Articles published in CDEI include original quantitative and qualitative research, scholarly reviews, and program descriptions and evaluations.

Complete author guidelines may be obtained from the online submission site: https://mc.manuscriptcentral.com/proed/cdei Go to the grey Resources box and select the Instructions & Forms link.