This study is an investigation of whether social skills training
provided by paraprofessionals to elementary grade children with Autism
Spectrum Disorders (ASD) in both partially and fully included classrooms
can result in perceived gains in social skills as measured by teacher
ratings. Results showed that several areas of social responsiveness
noticeably improved as a result of the intervention in the short run;
however, sustained improvement was difficult to detect. This study
extends the research on the development of social skills among children
with ASD by examining perceptions of social responsiveness rather than
noting how often the children engaged in prosocial behaviors. It further
extends the research by studying the efficacy of using trained
paraprofessionals to deliver the intervention inconspicuously in the
child's general education classroom.
Keywords: autism spectrum disorders; social skills interventions;
According to the U.S. Department of Education (2002), the number of
students with Autism Spectrum Disorders (ASD) receiving educational
services in special education increased over 1300%, from 5,415 to 78,749
children during the 1990's. This number can only continue to rise
as current estimates put the prevalence of autism at 1 in 150 children
(Autism Speaks). At the same time as the number of children with autism
requiring services in our schools is increasing, calls for
evidence-based practices for all children have strengthened (Williams,
Johnson, & Sukhodolsky, 2005). The use of evidence-based practices,
those having an empirical basis attesting to their validity and
efficacy, is one approach at accountability.
The needs of children with ASD are indeed varied and complex,
however, the lack of social skills may be among the most pressing of
problems. There are many issues surrounding social skills interventions,
including but not limited to who should deliver the intervention (a
trained peer or a trained adult), where the intervention should take
place (e.g., in the school, in the clinic, or in a pull-out program
within the school), and what the intervention should be (e.g., social
scripts, modeling, visual cueing, discrete trials). While studies using
trained peers show promising results, the question of acceptance of
children with ASD by their peers remains relevant and problematic.
Farmer, Pearl and, VanAcker (1996) noted that students with disabilities
are more likely to have a rejected status and less likely to have a
popular status in the classroom and school. The concept of peer
rejection is derived from peer nominations and peer ratings that measure
how well students are liked by classmates in general, and is based
heavily on the degree of a child's social skills (Farmer, Pearl,
& VanAcker). Behaviors such as the stereotyped behaviors often
common in ASD are not seen as socially acceptable and frequently reduce
opportunities for social interaction (Lee, Odom, & Loftin, 2007).
Chamberlain, Kasari, and Rotherman-Fuller (2007) found that "the
average level of social network centrality was lower for children with
autism than for their peers; they were less well accepted and they had
fewer reciprocal friendships" (p. 239). Swaim and Morgan (2001)
have shown that even with information explaining autism, attitudes of
typically developing children toward children with autism do not change.
The work of Lee, Yoo, and Bak (2003) shows that children with typical
abilities tend to act as helpers, caregivers, and tutors of children
with disabilities rather than becoming friends, as evidenced by
reciprocal social interactions. Though this is not the ideal outcome of
peer tutoring, it allows for a give and take relationship and may afford
the child with ASD some opportunity to move from weak social skills to
more appropriate social skills. Contact with typically developing peers
is thought to be a critical piece in supporting children with ASD, but
it cannot be the only piece representing successful socialization (Ochs,
Kremer-Sadlik, Solomon, & Gainer Sirota, 2001). The opportunity to
see their peer's model appropriate social skills does not
automatically translate into the ability to interact with their peers in
a meaningful manner (Constantino et al., 2003). Because children with
ASD have difficulty learning functional sequences, vicarious learning of
the functional sequences of social skills is problematic. Most often
such sequences cannot be completed without receiving direct instruction,
a visual prompt, or a specific routine from teachers, aides, or other
students (Mesibov & Shea, 1996). This draws into question the
efficacy of using peers as change agents for children with ASD.
Many of the characteristics of ASD are the cause of peer rejection.
Their social deficits stem from alack of understanding the importance of
social norms (Heery, Capps, Keltner, & Kring, 2005). In schools
there is a set of social norms, known as the hidden curriculum. These
are the set of rules that everyone in the school is expected to know,
but that no one has been directly taught (Heery, Capps, Keltner, &
Kring; Smith Myles, 2005). For those children who have difficulty with
picking up on the hidden curriculum, isolation and rejection are natural
consequences. When others perceive peers to act in accord with social
norms, acceptance is often forthcoming. It stands to reason then that
when others perceive that children with ASD engage in more socially
acceptable behaviors, or at least less odd social behaviors, these
children might enjoy greater social acceptance and less rejection. If
the skills one develops are judged as more socially acceptable (rather
than the frequency with which one actually performs such skills), then
that perception may be a critical factor in progress toward ultimate
social acceptance. Not learning social skills at a young age can have
long lasting implications for the ability to create relationships and to
interact appropriately as an adult (Chamberlain, Kasari, &
Successful social integration in the regular classroom emphasizes
being visible to other students (social impact), being someone with whom
other students want to interact (social preference) and being a member
of a group (social network affiliation) (Chamberlain, Kasari, &
Rotherman-Fuller, 2007). Classroom social networks are created by the
students as they develop distinct associations and determine the
behaviors acceptable within the social networks (Farmer, Pearl, &
VanAcker, 1996). Studies have shown that children with ASD are more
socially involved with their peers when in inclusive classrooms (Mesibov
& Shea, 1996; Williams, Johnson, & Sukhodolsky, 2005). However,
proximity alone to normally developing peers is unfortunately
insufficient for achieving social skills goals (Constantino et al.,
2003). Knowing this, the inclusive classroom would appear to be an ideal
place to implement formal and deliberate social skills interventions.
Social skills interventions carded out in an inclusive setting, where
the hidden curriculum occurs, might have more success in producing
measurable change than interventions carried out in pull-out programs
because of the availability of typically developing peers with whom to
interact, the regularity with which the intervention can occur, and the
fact that skills can be taught and practiced in the environment in which
they are targeted to occur (Ingersoll & Schreibman, 2006; Williams,
Johnson, & Sukhodolsky, 2005). But do teachers have the time and the
opportunities to manage the interventions?
Paraprofessionals provide valuable service for many children in the
classroom. They often implement interventions working one-on-one with
children for reasons such as reminders to stay on-task, to follow
directions, or to assist with scaffolding the teacher's whole group
instruction. However, despite the important position they play in many
children's lives, they are not always well-prepared for the roles
they are assigned (Malmgren, Causton-Theoharis, & Trezek, 2005).
Yet, they continue to be called on for assistance for educational and
behavioral goals for children in their care. Malmgren,
Causton-Theoharis, and Trezek suggested that training paraprofessionals
to teach social skills in the classroom can allow for these skills to be
taught in the inclusive classroom without disrupting the flow of
teaching or removing the child from opportunities for social interaction
with peers. This arrangement might also address the issue of transfer of
learned social skills across settings, something that is often a problem
for many children with autism.
Social skills interventions all rely on either an adult or a peer
to encourage appropriate social behaviors. Many use the natural setting,
for example recess and the play ground, while others use a more
contrived setting where role playing and script learning are practiced
before entering the natural setting. The majority of the research done
with social skills training among children with ASD demonstrates
effectiveness through the use of behavioral counts of actual prosocial
behavior prior to, during, and following the intervention. However, as
suggested previously, social acceptance depends as much on how the
behaviors of the child are perceived by others as on the frequency with
which the child demonstrates the skills.
The purpose of this study was to investigate the effectiveness of
the use of visual cueing for peer interaction and social skill
reminders, delivered by trained paraprofessionals in the general
education classroom, on teacher perceptions of the social responsiveness
of children with ASD. Specifically, the questions considered were: (1)
would paraprofessionals trained to deliver a set of interventions be
able to do so in the general education classroom? (2) would teachers be
able to detect changes in social responsiveness among their students
with ASD as a result? and (3) did the educational placement (partial or
full inclusion) have an effect on teacher's ratings of progress of
social responsiveness among these children? This study extends the
research on the development of social skills among children with ASD by
examining perceptions of social competence rather than noting how often
the children engaged in prosocial behaviors. Social competence is
considered a summary judgment term based on an individual's
performance on a given social task (Reschley & Gresham, 1981).
Therefore, how one's behaviors and methods of interaction are
perceived by others determines judgments of social competence. According
to Reschly and Gresham, judgments of social competence include the more
discrete and directly observable social skills. It further extends the
research by studying the effectiveness of using trained
paraprofessionals to inconspicuously deliver the intervention in the
child's general education classroom.
Materials and Method
Participants for this study were chosen under nonrandom selection.
Criteria for participation included a current diagnosis of autism made
by independent evaluators, a current IEP that included the use of a
paraprofessional, part or full-time inclusion in the general education
classroom, and enrollment in kindergarten through fifth grade. Ten
students fit the criteria and were asked to participate; seven
ultimately participated in the study. The students participating
included two first-graders, one second-grader, one third-grader, and
three fourth-graders. All of the children were accompanied by an aide
full-time during the school day. All children in the study were of
White, Non-Hispanic descent and male. The two children in the first
grade were partially included in the general education classroom. The
preferred mode of communication of both first graders was echolalic
communication and neither initiated interactions with their peers. The
second grade student was fully included in the general education
classroom. He communicated with adults and peers using words, but would
not initiate interactions with his peers. He was sought out by peers as
a good choice for teams in the classroom, but at recess would play near
rather than with his peers. The third grade student was partially
included in the general education classroom. This child had a tendency
toward perseverative thinking; did not respond to peers' attempts
to interact and did not initiate interactions with anyone unless he had
a physical need (i.e. snack, lunch, restroom). Of the fourth graders,
one student was very verbal and loved to initiate talking with adults,
but would not initiate interactions with peers though they would
sometimes initiate interaction. Understanding how or when to initiate
talking with adults was problematic and attempts often interfered with
lessons or other conversations. This student was fully included and used
the autism classroom as a homeroom. The other two fourth grade students
were less verbal and also fully included. Neither initiated interactions
with their peers or adults unless they had a physical need.
This study took place in five general education classrooms in one
public school. The intervention was carried out by seven trained
paraprofessionals. Social Skills Training: For Children and Adolescents
with Asperger Syndrome and Social Communication Problems ([SST] Baker,
2003) served as the intervention, from which 6 skills were identified to
serve as the focus for this study. The skills were introduced to the
student by the paraprofessional in the following order: Maintaining
Appropriate Physical Distance from Others, How and When to Interrupt,
Editing Sensitive Subjects, Recognizing Feelings, Dealing with Making
Mistakes, and Trying When Work is Hard. The paraprofessionals were
trained on the implementation of the focus social skills and visual cues
during a two-hour training which occurred one month prior to
implementation. This training demonstrated each focal social skill and
the visual cue that accompanied that social skill. Each paraprofessional
was instructed on how to introduce the visual cue, how to help the
student complete the social skill by using the visual cue, and how to
incorporate the use of the social skill and visual cue without
disturbing what was going on around the student. The paraprofessionals
practiced using the social skills and visual cues with each other during
the training .Adjustments on technique and implementation were made as
needed. Weekly 30-minute sessions were held during the month prior to
implementation to ensure that the paraprofessionals understood the
technique, to allow for practice, and to answer questions. Once the
intervention began, each paraprofessional was observed implementing the
social skills and visual cues with the student on the first day of
implementation for fifteen minutes by the trainer. Correction on
technique and implementation was provided where needed, along with
answering questions. Observation continued daily in 15 minute blocks
approximately every 90 minutes until it was determined that the
intervention was being implemented correctly. After two days the
rotating observations were no longer needed. Weekly 30-minute training
sessions continued during the implementation of the intervention for the
paraprofessionals to allow for answering of questions and refreshing
One new skill was introduced weekly by the paraprofessional along
with review of previously learned skill(s) as indicated by the SST
manual. Paraprofessionals introduced the skill by showing the new visual
cue card for the skill to the child, describing each symbol, and
role-playing a situation that fit the skill. The same was done with the
skill(s) being reviewed. The next time the skill was needed in social
situations, the card was shown to the child, verbally reviewed, and
practiced with the other person involved in the situation. As the child
mastered the skill, verbal reminders were faded and only the card was
shown to the child.
Social skills data was collected using the Social Responsiveness
Scales ([SRS], Constantino, 2005). The SRS is a 65-item rating scale
that measures the severity of autism spectrum symptoms as they occur in
a natural setting. The questionnaire inquires about a child's
ability to engage in emotionally appropriate reciprocal social
interactions. The SRS also assesses communicative deficits and
restricted/stereotypic behaviors or interests. The SRS has good internal
consistency (.93-.97) and validity (Constantino).
The SRS provides 5 subscale scores and a total social rating score.
The subscale scores are: "(a) social awareness or the ability to
pick up on social cues and reciprocal social behavior; (b) social
cognition or the ability to interpret social cues and
cognitive-interpretive aspects of reciprocal social behavior; (c) social
communication or expressive social\communication; (d) social motivation
or the extent to which a respondent is generally motivated to engage in
social-interpersonal behavior; and (e) autistic mannerisms which
includes stereotypical behaviors or highly restricted interests
characteristic of autism" (Constantino, 2005, p. 17). Each item on
the SRS scale inquires about an observable behavior that is rated on a
"0" (never true) to "3" (always true) Likert scale.
The scores are translated into a single total scale score that serves as
an index of the overall severity of social deficits. High scores on the
SRS indicate greater severity of social impairment.
The participants' general education teachers were asked to
complete the SRS three times during the study: at baseline, mid-way
through the study, and at the end of the study.
Nonparametric Friedman tests, with Wilcoxon Signed Rank tests as
follow-up tests, were conducted to examine the progress made by the
students. Measurements were taken at baseline, mid-intervention and end
of the intervention using the SRS, completed by the children's
teachers. Mann Whitney U tests were conducted to determine group effects
of partial and full inclusion on social skills changes.
Results and Discussion
Several interesting findings emerged from the analyses. First, Mann
Whitney U tests showed no significant differences between children who
were partially or fully included on social responsiveness scores. Table
1 contains the means and standard deviations of the subscales and total
social rating scores for children in the fully-included classrooms and
those in the partial-inclusion classrooms at baseline and end of
Friedman tests showed significant differences in several areas.
Ratings by teachers showed significant gains in four of the six
subscales and the total social rating score as measured by the SRS.
Social awareness scores differed significantly with Friedman's test
[chi square] (2) = 7.44, p = .024; social cognition scores differed
significantly with Friedman [chi square] (2) = 6.522, p = .038; autistic
mannerisms scores differed significantly with Friedman [chi square] (2)
= 7.154, p = .028; and total social ratings scores with Friedman [chi
square] (2) = 6.741, p = .034. Social motivation and social
communication were not significantly different with p =. 112 and .084,
respectively. Follow-up Wilcoxon Signed Rank tests showed significant
social awareness differences between baseline (M = 66.14, SD = 6.91) and
mid-intervention (M = 59.14, SD = 6.99); significant social cognition
differences between baseline (M= 69.14, SD = 9.86) and mid-intervention
(M = 63.43, SD = 7.61); significant autistic mannerisms differences
between baseline (M = 66.71, SD = 5.50) and mid-intervention (M = 63.14,
SD = 7.54); and significant total social ratings score differences
between baseline (M = 67.57, SD = 6.68) and mid-intervention (M = 59.71,
SD = 7.87). No other differences were significant. Figure 1 shows the
mean scores at baseline, mid-intervention, and end of intervention in
the subscales where significant changes were found.
The results of this study indicate that paraprofessionals who are
appropriately trained can help children with ASD to learn more
appropriate social skills; that the paraprofessional can intervene to
help the child make progress while the child is in the general education
classroom; and that the children's teachers perceive more
acceptable social skills on the part of the children. However, the
teachers rated the children more highly at the mid-point of the
intervention than at the end of the intervention, indicating perhaps
that the gains were not sustainable to the same degree over time. This
is both promising and discouraging. Lasting gains are the goal of any
intervention; and the results of this study do not point in the
direction of maintenance even over the short period of time of
six-weeks. However, an alternate explanation exists that might account
for the slight loss of ground in the areas where gains were noted
initially. Perhaps, the teachers' expectations drifted over time.
The ratings on the SRS did not revert to baseline levels but they did
increase (show less progress) from the mid-point. It might be that over
time the teachers began to expect more acceptable social skills because
the children were demonstrating more proficiency.
The children in this study were able to demonstrate social skills
improvement in a short period of time, leading one to conclude that the
use of immediate practice and prompting with visual cues can help a
child with ASD to be perceived as more socially adept within the general
education classroom. This result supports previous research that
indicated positive inclusive classroom social interactions are needed to
assist children in gaining social skills that are essential in
adolescence and adulthood (Chamberlain, Kasari, & Rotherman-Fuller,
2007; Mesibov & Shea, 1996; Sale & Carey, 1995).
[FIGURE 1 OMITTED]
The results also indicate that paraprofessionals can be effective
in providing intervention in social skills and that such intervention
can result in measurable gains. The use of a paraprofessional to
implement the intervention can allow for the integration of skills
immediately, as social issues occur in the classroom, and with immediate
feedback and practice. Perhaps it is because of the immediate feedback
and practice such rapid gains were seen. As noted by Siegel (2003),
children with autism don't have to be pulled from the class to be
taught social skills. It might be better to intervene in the more
authentic environment. Further, the results did not differ if the child
was educated in the general education classroom only or if the child was
educated in a partially-included arrangement. All children in the study,
whether fully or partially included, made improvements in social skill.
Limitations and Recommendations for Further Research
Aside from the issues of small sample size and the involvement of
only male students, the design of the study incorporated the social
skills at a rate of one per week. With this design the last social skill
taught had only one week of practice and therefore most likely not
enough time to show changes in social skills ability or integration of
that skill into the developing skill set.
Future research could improve on the current study by increasing
the sample size, including both male and female participants, and middle
and high school students with ASD. This study noted the teacher's
perception of social skills changes. Future studies might investigate
the perception of the child's peers relative to the development of
more prosocial behaviors.
In conclusion, significant gains were observed in a very short
period of time in the areas of social awareness, social cognition,
autistic mannerisms, and overall social responsiveness when social
skills training was implemented by paraprofessionals in the general
education classroom. Gains were consistently seen for both partially and
fully-included students. For children with ASD, the results suggest that
social skills training by paraprofessionals within the general education
classroom can create gains in social skills that are perceptible by
teachers and could lead to greater acceptance by their classroom peers.
Autism Speaks, Retrieved June 17, 2009, from
Baker, J. (2003). Social Skills Training: For Children and
Adolescents with Asperger Syndrome and Social-Communication Problems.
Western Psychological Services, Los Angeles, CA.
Chamberlain, B., Kasari, C., & Rotheram-Fuller, E. (2007).
Involvement or isolation? The social networks of children with autism in
regular classrooms. Journal of Autism and Developmental Disorder, 37,
Constantino, J. (2005). Social Responsiveness Scale. Western
Psychological Services, Los Angeles, CA.
Constantino, J., Davis, S., Todd, R., Schindler, M., Gross, M.,
Brophy, S., et al. (2003). Validation of a brief quantitative measure of
autistic traits: Comparison of the Social Responsiveness Scale with the
Autism Diagnostic Interview--Revised. Journal of Autism and
Developmental Disorders, 33 (4), 427-433.
Farmer, T, Pearl, R, & Van Acher, R. (1996). Expanding the
social skills deficit framework: A Developmental synthesis perspective,
classroom social networks, and implications for the social growth of
students with disabilities. The Journal of Special Education, 30,
Heerey, E., Capps, L., Keltner, D., & Kring, A. (2005).
Understanding teasing: Lessons from children with autism. Journal of
Abnormal Child Psychology, 33(1), 55-68.
Ingersoll, B. & Schreibman, L. (2006). Teaching reciprocal
imitation skills to young children with autism using a naturalistic
behavioral approach: Effects on language pretend play, and joint
attention. Journal of Autism and Developmental Disorders, 36(4),
Lee, S., Odom, S., & Loftin, R. (2007). Social engagement with
peers and stereotypic behaviors of children with autism. Journal of
Positive Behavior Interventions, 9(2), 67-79.
Lee, S., Yoo, S., & Bak, S. (2003). Characteristics of
friendships between children with and without mild disabilities.
Education and Training in Developmental Disabilities, 38(2), 157-166.
Malmgren, K., Causton-Theoharis, J., & Trezek, B. (2005).
Increasing peer interactions for students with behavioral disorders via
paraprofessional training. Behavioral Disorders, 31(1), 95-106.
Mesibov, G. & Shea, V. (1996). Full inclusion and students with
autism. Journal of Autism and Developmental Disorders, 26(3), 337-346.
Ochs, E., Kremer-Sadlik, T., Solomon, O., & Gainer Sirota, K.
(2001). Inclusion as social practice: Views of children with autism.
Social Development, 10(3), 399-419.
Reschly, D.J., & Gresham, F.M. (1981). Use of social competence
measures to facilitate parent and teacher involvement and nonbiased
assessment. Unpublished manuscript, Iowa State University, Ames.
Sale, P. & Carey, D. (1995). The sociometric status of students
with disabilities in a full inclusion school. Exceptional Children,
Siegel, B. (2003). Helping children with autism learn: Treatment
approaches for parents and professionals. New York: Oxford University
Smith Myles, B. (2005). Children and youth with Asperger Syndrome:
Strategies for success in inclusive settings. Thousand Oaks, CA: Corwin
Swaim K. & Morgan S. (2001). Children's attitudes and
behavioral intentions toward a peer with autistic behavior: Does a brief
educational intervention have an effect? Journal of Autism and
Developmental Disorders, 31(2), 195-205.
United States Department of Education (2002). 24th Annual Report.
Retrieved February 20, 2008 from
Williams, S., Johnson, C., & Sukhodolsky (2005). The role of
the school psychologist in the inclusive education of school-age
children with autism spectrum disorders. Journal of School Psychology,
Rebecca Mazurik-Charles, Berwick Area School District. Candice
Stefanou, Associate Professor, Bucknell University.
Correspondence concerning this article should be addressed to
Rebecca Mazurik-Charles at rcharles @berwicksd.org.
Means and Standard Deviations of Subscale and Total SRS Scores at
Baseline and End of Intervention Assessment of Social Skills
Full-Inclusion (N = 3)
Score M(SD) M(SD)
Social Awareness 67.3 (4.04) 56.7 (6.5)
Social Cognition 67.7 (7.4) 62.7 (11.7)
Social Communication 65.7 (6.0) 60.0 (6.9)
Social Motivation 66.0 (9.6) 54 (3.6)
Autistic Mannerisms 70.0 (3.7) 65.3 (5.7)
Total Social Rating 68.7 (4.7) 60.3 (4.6)
(N = 4)
Score M(SD) M(SD)
Social Awareness 65.2 (9.1) 3.7 (5.7)
Social Cognition 70.2 (12.5) 8.7 (11.3
Social Communication 63.5 (6.7) 3.0 (6.5)
Social Motivation 66.7 (10.7) 0.2 (5.1)
Autistic Mannerisms 64.2 (5.7) 4.7 (8.5)
Total Social Rating 66.7 (8.5) 4.7 (8.0)