The Social Face of Inclusive Education: Are Students with Learning Disabilities Really Included in the Classroom?

The movement toward inclusion of students with disabilities into general education classes has become the overwhelming trend in education (Chow & Kasari, 1999; Mamlin, 1999). Not only does inclusive education for children with disabilities bring improved academic functioning (Manset & Semmel, 1997; Sideridis et al., 1997), but it also offers them the opportunity for socialization with their peers without disabilities in general education classrooms (Giangreco, Dennis, Cloninger, Edelman, & Shattman, 1993; National Center for Educational Restructuring and Inclusion, 1994). While early studies have investigated the academic performance of children with disabilities in inclusive settings, there has been increased interest in and attention to the social adjustment and social functioning of children with disabilities in inclusive settings (Vaughn, Elbaum, & Schumm, 1996; Vaughn, Elbaum, Schumm, & Hughes, 1998).

Students with disabilities often demonstrate delays in social development that parallel delays in their academic performance and achievement (Odom, McConnell, & Chandler, 1994). Some students lack skills in initiating and sustaining positive social relationships (Gresham, 1997; Heiman & Margalit, 1998) and in appropriately interpreting social cues (Heron & Harris, 1993). They often exhibit more aggressive and negative verbal and nonverbal behaviors (McConaughly, Mattison, & Peterson, 1994; Sigafoos, 1995) and may be either disruptive or withdrawn (Clare & Leach, 1991; McIntosh, Vaughn, & Zaragoza, 199 1). Often these behaviors result in students with disabilities having fewer friends than their peers without disabilities as well as their being actively rejected by peers (Farmer & Rodkin, 1996; Nabasoku & Smith, 1993). Such pervasive deficits in social functioning manifested by many students with learning disabilities have been widely acknowledged by the special education community. In fact, in 1987, the Interagency Committee on Learning Disabilities proposed modification of the definition of learning disabilities to include social skill deficits as a primary learning disability.

Social status can take a variety of forms. For example, Coie, Dodge, and Coppotelli (1982) devised a classification system by which they identify popular, rejected, neglected, controversial, and average children. Popular children are those children who, on a sociometric peer nomination instrument, receive a large number of positive choices and few, if any, negative choices. Such children are desired by peers. Conversely, rejected children receive a greater number of negative nominations but few, if any, positive ones. Neglected children are those persons who receive few positive or negative nominations. They are actively ignored, almost invisible persons in social settings. One investigator (Luftig, 1999) has taken to calling them “ghost children” because of their social invisibility.

Controversial children are those who receive a significant number of both positive and negative peer nominations. They are children who appear to be both positive and negative at the same time to peers. Finally, average children receive about the average number of both positive and negative nominations when compared with their peers.

Contradictory data exist regarding the effects of inclusive education settings on the social functioning of students with disabilities. For instance, Vaughn and others (1996) found inclusive classrooms to have a positive impact on the peer relationships and self-concept of students with learning disabilities. These students reported increases in reciprocal friendships and lower levels of social alienation after being in an inclusive classroom for the entire school year. Furthermore, in analyzing the social networks within inclusive third- and fourth-grade classrooms, T. Farmer and E. Farmer (1996) found that students with and without disabilities belonged to a peer cluster at about equal rates. Other researchers, however, report that being placed in an inclusive setting may pose additional difficulties for students with disabilities who demonstrate deficits in social behavior. Some evidence indicates that students without disabilities as well as teachers in general education classrooms often do not accept a student with disabilities (Bryan, 1997; Sale & Carey, 1995). Peers and teachers often ignore or actively reject the overtures of such students, praise them less, and consider them less desirable than students without disabilities (Heron & Harris, 1993; Stitt et al., 1988).

If students with disabilities are more isolated and/or rejected than their peers without disabilities, the question arises as to how acutely students with disabilities perceive and internalize such feelings of rejection when they occur. Put another way, when these students are actively rejected or ignored by their peers without disabilities, do they perceive and take to heart such rejection and what does that rejection make them feel about their future ability to initiate or sustain appropriate social relationships?

Students’ perceptions of rejection or of being ignored and their resultant feelings of isolation and disconnection often are manifested in feelings of perceived loneliness (Eronen & Nurmi, 1999; Sermat, 1980). Feelings of loneliness among children without disabilities have been found to range from 10% (Asher, Hymel, & Renshaw, 1984) to as high as 18% (Luftig, 1985). By comparison, rates of loneliness felt among children with developmental disabilities have been found to be much higher, often 25 % or more (Luftig, 1987).

Being rejected or ignored by peers may also affect one’s beliefs about the ability to make and keep friends in the future. This construct has often been referred to as “social competence” (Antia & Kreimeyer, 1992; Harter & Pike, 1984). Students’ impressions about their social competence influence their beliefs about the probability that future social initiations will succeed or fail (Parker & Asher, 1987; Nurmi, Toivonen, Salmela-Aro, & Eronen, 1996). Such beliefs are important for an individual making a decision to seek friendships and other positive social relationships in the future or to become even more isolated or withdrawn (Elias, 1995; Newton, Taylor, & Wilson, 1996).

The topic investigated in the present study was the perceived loneliness, social competence, and social status of children with learning disabilities in sixth-grade inclusive education settings. In this particular study, classrooms were considered inclusive when students with learning disabilities spent 100% of the school day in the general education classroom with same-age peers. Instruments that measured the 3 above-mentioned variables were administered to students with learning disabilities and to their classmates without disabilities. It was hypothesized that inclusive education would have a positive effect on the social functioning of students with learning disabilities and that thus they would be as accepted as their peers without disabilities. Consequently, the study would show no differences between the groups’ perceived loneliness and social competence.

Methodology

Participants

Participants in the study consisted of 15 students with learning disabilities and 68 students without disabilities, all of whom were enrolled in sixth-grade classrooms that had adopted a full inclusion model. All participants were from a single urban school district in southwest Ohio that had approximately 7,000 pupils. The school district policy was to educate all students with learning disabilities in the general education classroom with same-age peers for the entire school day. Students received academic assistance from the special education teacher and other support personnel in the form of co-teaching or small-group instruction within the general education classroom.

Students participating in the study were from 4 sixth-grade classrooms in three elementary schools in the district, each with an enrollment greater than 480 students. At least 40% of the students in each school were receiving free or reduced lunch Only students who brought back signed parental permission slips indicating consent to participate in the study were included. Data collected from students in the participating classrooms who had a diagnosed disability other than a learning disability were not included in the present analysis. No a priori decisions were made regarding the gender, ethnicity, socioeconomic status, or language background of the participants. Due to issues of privacy and confidentiality, researchers did not obtain information regarding the intelligence or achievement levels of participating students, other than that which ascertained their special education status and label.

Students with learning disabilities were included in the study if they (a) met the state of Ohio and district criteria (consistent with federal regulations) for a learning disability and (b) were fully included in a sixth-grade general education classroom for 100% of the school day. Of the participating students with learning disabilities, 8 were boys and 7 were girls. Their ages ranged from 10 to 13 years, with a mean chronological age of 11 years. All students were native English speakers. Participants had diagnosed learning disabilities in the areas of reading, writing, and math. Of the 68 sixth graders without disabilities, 36 were boys and 32 were girls. Students without disabilities were also between 10 and 13 years of age, with a mean chronological age of approximately 11 years.

Procedure and instruments

The study was conducted approximately 2 months after the start of the school year so that students had time to get to know one another and become familiar with the school and classroom environment. Two measures were administered to all participating students in their classroom by a proctor who was a trained school psychologist working in the school district. The proctor first introduced herself and the purpose of the study, namely, to learn more about how children feel about working and playing with other children at school. Students were informed that their participation in the study was voluntary and that they could stop participating at any time. They also were assured of the confidentiality of their responses and were instructed not to discuss their responses with others.

The two measures used in the study were the Modified Children’s Loneliness Scale (Luftig, 1986) and the Peer Nomination Inventory (Luftig, 1986). The measures were administered in a counterbalanced order, with the Loneliness Scale being administered first in two of the classrooms and the Peer Nomination Inventory being administered first in the other two classrooms.

The Modified Children’s Loneliness Scale was administered orally to each class by the proctor. Four practice items were included at the beginning of the questionnaire to ensure the participant’s familiarity with responding to the items on the measure. The proctor assisted students with these practice items. She then read aloud each item on the scale and the five response options that followed, and students read along silently, marking an X in the box representing the answer that was most true for them. Response options included, “that’s always true about me,” “that’s true about me most of the time”, “that’s sometimes true about me,” “that’s hardly ever true about me,” and “that’s not true at all about me.” Students were allowed to ask clarifying questions of the proctor after raising their hands, and the proctor quietly assisted them. This measure took about 30 min to administer.

Asher’s original scale has been found to demonstrate sound technical properties. For example, it possesses a Spearman- Brown reliability coefficient of 0.91 as reported by Asher and others (1984). Luftig (1987) also demonstrated concurrent validity of the scale to be 0.81 between teacher ratings and actual student reports of loneliness. An adaptation of the Children’s Loneliness Scale also has been used with 20 fourth- and fifth grade students with learning disabilities and has been found to have high reliability (coefficient a = 0.885) (Pavri & Monda-Amaya, 2000).

Students were seated in a semicircle so they could see all their classmates. The students were instructed to look around the class and nominate peers who best fit the given behavioral descriptions by writing their names on the inventory following each item. If a child was absent on that particular day, his or her name was written on the blackboard so students remembered to include that child in their nominations. The proctor read the items aloud to the whole class, and students were given time to respond to the item before the next item was read. If they needed to, students raised their hands to seek the assistance of the proctor in writing or spelling a peer’s name or in receiving clarification about the task, and the proctor quietly assisted them.

The items on the Peer Nomination Inventory formed two scales: “liked most” (popularity scale) and “liked least” (unpopularity scale) that were used to generate social impact and social preference scores. Of the 15 items, 8 items made up the “liked-most” scale and the remaining 7 items made up the “liked-least” scale.

Results

Data preparation and Scoring

Two scores were obtained for the Modified Children’s Loneliness Scale: a loneliness score and an estimate of perceived social competence. For each of the two dimensions measured by this scale, individual item scores were summed to yield a total loneliness and total perceived social competence score.

The order of some of the items was reversed so that all items were positively worded. Student responses were weighted and scored 1-5 so that a score of 5 always indicated greater loneliness or greater perceived social inadequacy. Means and standard deviations were computed for each score, which were used to conduct analyses of variance (ANOVAs) to determine whether there were differences between students with and without disabilities on these variables.

The responses on the peer nomination inventory were scored by tallying the number of nominations each student received on each item on the scale from all of his or her class peers. Coie and Kupersmidt’s (1983) procedure was then used to compute the student’s sociometric status. Based on peer nominations, each student received a total popularity and unpopularity score. This score was determined by summing the total score the student obtained from peers on the 8 questions representing popularity and the 7 questions representing unpopularity. Standard scores (z-scores) were then computed for each child to facilitate further statistical analyses. “Liked-most” (i.e., popularity) and “liked-least” (i.e., unpopularity) items were used to generate social preference and social impact scores.

The social preference score was the z-score (liked most) - the z-score (liked least), whereas the social impact score was the z-score (liked most) + the z-score (liked least). These social preference and social impact scores were used to define four extreme social status types. The popular group comprised those children who received a social preference z-score greater than +1.0, a liked-most z-score greater than 0, and a liked-least z-score less than 0. The rejected group comprised those children who scored a social preference z-score of less than -1.00, a liked-least z-score greater than 0, and a liked-most z-score less than 0. The neglected group included those children who received a social impact z-score of less than -1.00 and a liked-most and a liked-least z-score less than 0. The rejected children, however, received many more liked-least nominations than the neglected children did. The controversial group contained children who obtained a social impact z-score greater than +1.0 and liked-most and liked-least z-scores greater than 0. Last, the average group comprised children who received a social preference z-score greater than -1.0 and less than +1.0. Chisquare statistics were used to determine whether there were differences in the sociometric status of students with learning disabilities and their peers without disabilities.

Design and Data Analysis

The study employed a single factorial design with the independent variable being whether or not a child had a diagnosed learning disability. The dependent variables were the sociometric status (popular, rejected, neglected, or controversial), the perceived social competence, and perceived loneliness.

The data collected using the modified Children’s Loneliness Scalewere analyzed using a series of one-way ANOVAs to determine if there were differences between the students with learning disabilities and their peers without disabilities in perceived social competence and perceived loneliness. Significant differences were found between the two groups in perceived loneliness F(1, 55)=4.77, p ‹.03), but not in perceived social competence. Students with learning disabilities perceived themselves as being lonelier than their peers without disabilities. Table 1 shows the mean difference and standard deviations between the two groups on perceived loneliness.

As described earlier, the children were classified as belonging to 1 of 5 social status groups, depending on their social preference and social impact scores. Thus, the 15 students with learning disabilities were assigned to different social status groups. For this reason, there were not enough students with learning disabilities in each of the social status groups to yield statistical power and warrant parametric statistical analyses. Thus, nonparametric, chi-square analyses were used.

The central question regarding the social status data was whether the students with learning disabilities would be assigned to social status groups differently than was true for their peers without disabilities. Thus, the expected frequency or norm for students with learning disabilities in each social status group was the same as the frequency of the students without disabilities in that status group. Chi-square analyses were conducted to determine whether there were differences in the number of students with learning disabilities who were assigned to the popular, average, controversial, negative, and rejected social status groups as compared with the number of students without disabilities who were nominated to each of these status groups.

A significant chi-square was found for the social status groups of popular children and controversial children. Students with learning disabilities were less likely to be included in the popular group (df = 1, x2 = 8. 10, p (df = 1, x2 = 4.86, P ‹ .05) than their peers without disabilities.

Discussion

This study investigated whether students with learning disabilities who were educated in inclusive general education classrooms differed from their same-age peers without disabilities on the variables of social status and/or perceived loneliness. The results indicated that sixth-grade students with learning disabilities reported more feelings of loneliness than their classroom peers who did not have disabilities. Furthermore, these same students with learning disabilities were less popular and more controversial in their social status than their classmates without disabilities.

The finding of higher levels of reported loneliness among students with learning disabilities is interesting and to some extent consistent with the existing literature (Asher & Gazelle, 1999; Guay, Boivan, & Hodges, 1999). Students with disabilities in general have been found to report higher levels of perceived loneliness than their peers without disabilities (Luftig, 1987; Margalit, 1998). This finding has held true for students with diverse disabilities, including developmental disabilities or mental retardation (Luftig, 1988), students with physical disabilities (King, Specht, Schultz, & Warr-Leeper, 1997), gifted students (Kaiser & Berndt, 1985; Kline & Short, 1991), and students with learning disabilities (Coleman, McHam, & Minnett, 1992; Vaughn, Elbaum, & Schumm, 1996).

Loneliness is a perceived phenomenon. That is, people may feel lonely if they are truly rejected by peers or if they do not adequately perceive or understand their actual popularity among peers. Thus, of importance is why these students perceive themselves to be lonely. Put another way, the question may be asked as to how realistic or versed in reality are their feelings of loneliness.

Two possible explanations exist. The first explanation is that students with learning disabilities are actually disliked or ignored by peers and that thus their feelings of loneliness are realistic. The second explanation is that the social relationships of students with learning disabilities do not differ from those of their peers without disabilities, and that thus their feelings of increased loneliness are not grounded in reality and are largely a misconception on their part.

In the present study, it appears that the loneliness of the students with learning disabilities was realistic and related to their diminished social status. The findings suggest that the students with learning disabilities were less likely to be popular than their peers without disabilities and thus less likely to be nominated for social activities by peers. Given such a lack of nominations for social activities, it is not surprising that the students with learning disabilities were aware of their social isolation and described themselves as lonely.

Another new finding from the present study was the increased likelihood of students with learning disabilities to achieve the controversial social status, a situation where a student achieves a significant number of both positive and negative nominations. Past studies have shown the controversial category to be relatively small among students (Coie & Dodge, 1983), with the total percentage of students falling in this category being about 5%. In the present study, about 7% of the students without disabilities fell into this category, whereas more than 13% of the students with learning disabilities were classified as having controversial social status.

Why did a higher percentage of students with learning disabilities fall into the controversial category? By definition, these students were engaging in certain behaviors that caused them to be unpopular and popular at the same time or, conversely, to be popular with some students but unpopular with others. A number of researchers have found that students with learning disabilities show decreased social acceptance by their peers without disabilities (Asher & Taylor, 1981; Stanovich, Jordan, & Perot, 1998) and that these students often are rejected by peers due to aggressive or inappropriate social skills (Bryan, 1997; Heron & Harris, 1993). Yet the present investigators found no studies that reveal significant differences in the rate at which students with disabilities are nominated for the controversial category.

It would have been beneficial to identify the specific behaviors that the students with learning disabilities engaged in that resulted in their higher incidence in the controversial category. This information is important because evidence shows that students’ behaviors are often a more accurate indicator of their social status than the label placed on them for the purpose of providing special education services (Raymond & Matson, 1989; Roberts & Zubrick, 1993). Additional investigation is required regarding the specific behaviors that contribute to the inclusion of students with learning disabilities into the controversial category.

One interesting finding of the present study was that although students with learning disabilities were less popular and more lonely than their classmates without disabilities, they did not label themselves as being less socially competent. This finding is consistent with earlier studies that found that although students with learning disabilities were in reality less socially competent than their peers without disabilities (Coleman et al., 1992), they were also less accurate than their peers in assessing their own social status and competence (LaGreca & Stone, 1990). Their inaccuracy may occur for a variety of reasons relating to both development and cognitive functioning. For example, Vaughn, Hogan, Kouzekanani, and Shapiro (1990) found developmental differences in the accuracy of perceived social competence in students with learning disabilities, with older students being more accurate than younger students. However, Vaughn and others (1990) found that even students as young as kindergarten or first grade could begin to accurately assess their own social competence.

The students in the present study were enrolled in sixth grade and were presumably mature enough to accurately assess their social competence. Nevertheless, the students with learning disabilities assessed themselves to be as socially competent as their peers without disabilities even though they were decidedly less popular. Students with learning disabilities may demonstrate a cognitive social deficit (or social metacognitive deficit) that affects their social perceptions much as their cognitive learning deficits affect their ability to learn academic material (Bruck, 1986; Luftig, 1987). Such a metacognitive deficit would hinder their ability to adequately interpret feedback from others.

Finally, it should be pointed out that the current findings as to the decreased popularity and increased controversial status of the students with learning disabilities occurred within the inclusive education setting. A variety of studies have found that students often do not accept their peers who have disabilities (Bryan, 1997) and that they are more likely to reject or to be critical of the behaviors of such students (Stitt et al., 1988). Thus, it is important to note that merely placing students with disabilities in inclusive classrooms is not sufficient to allow for their social inclusion and that other supports need to be in place to facilitate their acceptance and belonging in the peer group.

In summary, it appears that students with learning disabilities were lonelier than students without disabilities and that their loneliness appeared to be versed in reality inasmuch as they were less popular and more controversial than their peers. Although further research on the specific behaviors that give rise to such isolation seems appropriate, it may be wise for classroom teachers to deal with student feelings of loneliness and depression while also teaching them required social skills.

Limitations of the study

There were two limitations to the present study that pose problems regarding generalization to other student populations. The first was the relatively small number of students with learning disabilities participating in this study. These results must be duplicated with similar students in other schools and with students at different developmental levels before they can be generalized. The second limitation was the model of inclusive education adopted by the school system in which the current students were enrolled, which may be different from the models of inclusive education adopted by other schools. The school followed what they called the “full inclusion” model, in which all students with learning disabilities were served in the general education classroom for the entire day. It would have been beneficial to determine what types of social support were available to students with disabilities to facilitate their social functioning and peer relationships in the inclusive education setting.

Implications for practitioners

As stated earlier, it appears that merely placing students with disabilities in inclusive classrooms is not sufficient to allow for their social inclusion, and other supports need to be in place to facilitate their acceptance by and belonging to the peer group. One type of support needed for acceptance of students with disabilities in inclusive classrooms is teacher support (Chow, 1999). Research evidence indicates that teachers’ attitudes toward students with disabilities in inclusive classrooms are the most powerful factors in these students being accepted by peers, reporting less loneliness, and maintaining their self-esteem (Campbell, Dodson, & Bost, 1985; Luftig, 1985). Some general education teachers have been found to be anxious about or even resistant to having students with disabilities placed in their classes (Fiedler & Simpson, 1987; Gerstein & Luftig, 1998). There is a clear need for teacher preparation programs to address the social acceptance of students with disabilities in the general education classroom and to provide teachers with strategies needed to facilitate the social functioning of all their students.

This does not imply, however, that the onus for the social acceptance of students with disabilities is entirely on the teacher. Regular classroom teachers and classmates without disabilities should not and cannot accept total responsibility for the social acceptance of peers with learning disabilities. That is, social acceptance is not automatic and is usually based on the student’s set of social behaviors that he or she demonstrates with peers. For this reason, it is important that the student with learning disabilities receive intentional and active coaching in learning the social behaviors that lead to acceptance (D. W. Johnson & R. T. Johnson, 1989; Searcy, 1996). A number of viable programs and methods exist to help students to achieve these goals (e.g., Meisgeier, 198 1; Zirpoli & Melloy, 1997), and teachers working with students with learning disabilities would be strongly advised to become competent and willing to provide such instruction.

About the authors

Shireen Pavri is an assistant professor of special education at California State University, Long Beach.

Richard Luftig is a professor of educational psychology at Miami University in Ohio.