ONLINE DELIVERY OF A SOCIAL SKILLS INTERVENTION WITH VIDEO MODELING
FOR ADOLESCENTS WITH AUTISM SPECTRUM DISORDER
By
Carolyn Anne Kuehnel

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
School Psychology – Doctor of Philosophy
2013

ABSTRACT
ONLINE DELIVERY OF A SOCIAL SKILLS INTERVENTION WITH VIDEO MODELING
FOR ADOLESCENTS WITH AUTISM SPECTRUM DISORDER
By
Carolyn Anne Kuehnel
As the number of children and adolescents diagnosed with autism spectrum disorder
(ASD) increases, researchers and practitioners are faced with the challenge of identifying
efficient and effective treatments to meet the needs of a large number of individuals. Various
interventions have been used with individuals with ASD to remediate social difficulties, one of
the primary areas of impairment for those with ASD. Group and single-case studies have
demonstrated positive effects through individual, peer-mediated, and group social skills
interventions for various age groups. The research largely focuses on early childhood however,
and the majority of interventions take place in a face-to-face context. This study examined the
effectiveness of an online video modeling intervention for adolescents in an effort to explore
online delivery as a novel medium for the delivery of social skills intervention for individuals
with ASD. To gather information about relevant intervention factors, participant feedback was
sought and attitudes about the intervention were assessed. In addition, this study examined the
relationships between social skills, mental health problems, and academic achievement for
adolescents with ASD, as well as the effects of social skill improvement on mental health
problems.
Fifty adolescents with ASD in grades 9-12 were included in this randomized controlled
study, as well as one parent of each participating adolescent. Adolescents were randomly
assigned to either a 6-week online social skills intervention group or a wait-list control group.
Data from all participants at pre-test were used to examine the relationships between social,

mental health, and academic variables, whereas changes from pre-test to post-test were compared
between groups to examine intervention effects. Additionally, this study examined reporter
effects by comparing adolescent and parent measures of social skills within each of the broader
research goals.
No significant effects of treatment on social skills were found when examining
adolescent and parent reports. A significant treatment effect was found, however, on parent
ratings of adolescent problem behavior and parent ratings of ASD symptoms. Parents and
adolescents had significantly different reports of social skills at pre-test and post-test, but a
comparison of gain scores indicated non-significant differences between groups in parent and
adolescent reports of social skills change over time. Perceived competence for learning,
interest/enjoyment, and attitudes toward using technology were not predictive of intervention
outcomes. Important relationships were found between social, mental health, and academic
variables, including significant predictive effects of parent social skills ratings on GPA, grades,
depression, and anxiety. Although adolescent social skill ratings had no predictive effects,
adolescent problem behavior ratings were predictive of anxiety and depression ratings.
Adolescent ratings of social skills change over time were predictive of changes in anxiety ratings
over time, but were not predictive of changes in depression ratings over time. This study
highlights implications for alternative delivery of interventions and provides a foundation for
future research on creatively and effectively serving all students with ASD who face social
challenges.

To my beloved Nana, whose love and strength inspired me.

iv

ACKNOWLEDGMENTS
It is a pleasure to thank those who made this dissertation possible. First and foremost, I
must thank my advisor and dissertation chair, Dr. Evelyn Oka, for her guidance and
encouragement over the past five years. Not only have you helped me to grow as a researcher
and practitioner, but your support for my interests and ideas has also helped me to find my
professional identity. More than anything, I am thankful for your helpful and enthusiastic
mentorship approach. You pushed me to explore topics I was passionate about, challenged me to
think critically, and actively supported me every step of the way. I also must express my
appreciation for the members of my dissertation committee, Dr. Cary Roseth, Dr. Joshua
Plavnick, and Dr. Esther Onaga. Each of you has been invaluable throughout the dissertation
process. Your careful guidance and thoughtful critiques have undoubtedly made this project
stronger.
I would also like to thank my family. PK, if it weren’t for you, I may not have ever
attended graduate school. You have often believed in me more than I have believed in myself,
and without your encouragement, I truly would not be where I am today. Thank you for
everything during these difficult five years, especially your tolerance for the demands of
graduate school. I look forward to our post-graduate years together. Mom, you have never
hesitated to show the world how proud you are of your kids. Thank you for your support and
praise. Thank you also for guiding me to my career path and teaching me how important it is to
care for others. Dad, your genuine interest in my work has meant the world to me. Thank you
for always expressing how much you care about my work, my happiness, and my future. Also,
thanks for the pancakes. Chris, thank you for being my big brother, always, near, or far. You
have taught me so much and I am thankful to have you. To Claire, Paul, and Dixie, thank you

v

for your constant support and for welcoming me into your family. To Rhett, thank you for
volunteering your help with this project.
I will be eternally grateful for all the friends who helped me make it through these past
few years successfully. To Candace, thank you for being the best friend I could ever ask for.
From study nights in undergrad and grad school applications to long-distance phone calls,
impromptu case conferences, camping trips, and wedding days – you’ve simply been there for
me through it all. To my far-away friends, Jenn, Katie, Evan, and Derek, thank you for taking
the time, energy, and effort to keep in touch and support me from afar while I pursued this
degree. I appreciate you more than I can say. To the Buddies team, thank you for the smiles,
laughter, and cheers. Special thanks to Eric, Mandy, Caleb, Paul, Justin, Katie, and Swifty, for
giving me a reason to fall in love with Michigan. I never expected to make such wonderful
friends during graduate school, but coffee shop writing parties, early morning tailgates, late night
games of ninja, nachos, dancing, and campfires with all of you are moments I miss terribly and
will never forget. To my Thursday night TV buddies, Kristen, Kayla, Jess, and Erin, thank you
for your support, sincerity, and friendship. Relaxing with you all over hot chocolate was always
a highlight of my week. Finally, to my internship buddy, Sarah, I express my deepest thanks.
When we first discovered our parallel lives, I was so excited to share internship year with
someone I could relate to. Little did I know that having you there to face this year’s challenges
would be so meaningful. Thank you for the pep talks, the commiserating, the Starbucks dates,
and most importantly, the chocolate.
This research was supported in part by grants from the Society for the Study of School
Psychology, the Michigan State University Graduate School, and the Michigan State University
College of Education.

vi

TABLE OF CONTENTS

LIST OF TABLES

x

LIST OF FIGURES

xii

KEY TO ABBREVIATIONS

xiii

CHAPTER 1 Introduction
Challenges for Adolescents with ASD
Purpose of the Study
Theoretical Framework
Evidence-based intervention
Social learning theories
Contextual development

1
3
8
11
12
13
14

CHAPTER 2 Review of the Literature
Social Skills Problems of Individuals with ASD
Psychological Adjustment Problems for Individuals with ASD
Success of Previous Interventions
Group interventions
Individual interventions
Interventions with peers
Video modeling interventions
Factors That Affect Intervention Success
Key Factors for Online Intervention Development and Implementation
Video modeling intervention procedures
Rationale for online interventions
Proposed structure for online implementation of video modeling
Current Study Research Questions and Hypotheses
Research Question 1
Research Question 2
Research Question 3
Research Question 4A
Research Question 4B
Research Question 5

16
16
21
23
24
29
32
34
35
36
36
38
42
44
44
45
45
45
45
45

CHAPTER 3 Research Methods
Pre-Dissertation Study
Methods
Results
Social Skills Intervention Development
Video model development
Website development

46
46
46
47
48
48
50

vii

Intervention revisions
Rationale for intervention components
Final considerations
Recruitment
Participants
Measures and Instruments
Social skills
Social Skills Improvement System
Autism Social Skills Profile
Difficulty Likert scales
Perceived social support
Depression
Anxiety
Demographic information
Attitudes
Intrinsic Motivation Inventory
Attitude Toward Using Technology Scale
Exposure
Participant satisfaction
Procedure
Time 1 assessment
Implementation
Time 2 assessment

51
53
54
55
56
62
62
62
63
64
65
66
66
66
67
67
67
68
68
68
68
69
70

CHAPTER 4 Results
Intervention Metadata
Preliminary Data Checks
Research Question 1: Does an online social skills intervention improve social
skills?
Research Question 2: How do parents and adolescents in the intervention
group compare in their views of social skill improvement?
Research Question 3: How do attitudes about the intervention relate to
intervention outcomes?
Research Question 4A: Are social skills and perceptions of social support
related to academic achievement or mental health problems?
Research Question 4B: Are these relationships different depending on
adolescent- or parent-report?
Research Question 5: Are adolescents’ self-reported changes in social skills
related to changes in mental health problems?

71
71
72

CHAPTER 5 Discussion
Efficacy of the Online Intervention
Conclusions
Possible explanatory factors
Measurement
Dosage and integrity

102
102
102
103
104
105

viii

73
82
88
91
99
100

Sample heterogeneity
Practical significance
Perspectives on Effectiveness
Conclusions
Influence of Attitudes on Intervention Outcomes
Relationships Between Social, Academic, and Emotional Variables
Regression models at Time 1
Regression models of change over time
Conclusions
Limitations
Future Research
Implications for Practice
Conclusions

105
107
107
109
111
111
112
112
113
114
116
120
121

APPENDICES
APPENDIX A Recruitment Letter
APPENDIX B Parent Informed Consent
APPENDIX C Adolescent Assent Form
APPENDIX D Demographic Questionnaire
APPENDIX E Parent Likert Scale Form
APPENDIX F Adolescent Likert Scale Form
APPENDIX G Parent Multidimensional Scale of Perceived Social Support
APPENDIX H Adolescent Multidimensional Scale of Perceived Social Support
APPENDIX I Intrinsic Motivation Inventory Perceived Competence for Learning
APPENDIX J Intrinsic Motivation Inventory Interest/Enjoyment
APPENDIX K Attitude Toward Using Knoodle Questionnaire
APPENDIX L Diagnostic Criteria154

123
124
127
132
136
140
142
144
146
148
150
152

REFERENCES

147

ix

LIST OF TABLES

Table 1: Krasny et al.’s (2003) Essential Ingredients of Group Social Skills Curricula

39

Table 2: Weekly Lesson Topics Adapted From Laugeson & Frankel (2010)

49

Table 3: Elements of Online Intervention

52

Table 4: Demographic Characteristics of Adolescent Participants

60

Table 5: School Demographic Information for Adolescent Participants

60

Table 6: Diagnostic Information of Adolescent Participants

61

Table 7: Intervention Information for Adolescent Participants

61

Table 8: SSIS Problem Behavior Subscale Raw Scores Corresponding to
Behavior Levels

63

Table 9: Current Study Scale Reliability

64

Table 10: Intervention Metadata

72

Table 11: Treatment Effects

81

Table 12: Pearson Correlations Between Pre-Test Measures

92

Table 13: Pearson Correlations Between Post-Test Measures

93

Table 14: Means and Standard Deviations for Social Support, Depression, and Anxiety

94

Table 15: Social Skills & GPA – Parent Ratings

95

Table 16: Social Skills & School Grades – Parent Ratings

96

Table 17: Prediction of Time 1 Depression Scores from Time 1 Parent Social Skills
Ratings

97

Table 18: Prediction of Time 2 Depression Scores from Time 1 Parent Social Skills
Ratings

97

Table 19: Prediction of Time 1 Anxiety Scores from Time 1 Parent Social Skills
Ratings

98

x

Table 20: Prediction of Time 2 Anxiety Scores from Time 1 Parent Social Skills
Ratings

98

Table 21: Comparison of Parent-Adolescent Regression Weights

100

Table 22: Social Skills and Anxiety – Adolescent Report

101

Table 23: Changes in Parent-Reported SSIS Descriptive Categories

107

Table 24: Diagnostic Criteria for Autistic Disorder

155

Table 25: Diagnostic Criteria for Rett’s Disorder

156

Table 26: Diagnostic Criteria for Childhood Disintegrative Disorder

157

Table 27: Diagnostic Criteria for Asperger’s Disorder

158

xi

LIST OF FIGURES

Figure 1: Mean Social Skills Scores at Time 2 from Parent Report

74

Figure 2: Mean Social Skills Scores at Time 2 from Adolescent Report

75

Figure 3: Mean Problem Behavior Subscale Scores at Time 2 from Parent Report

77

Figure 4: Mean Problem Behavior Subscale Scores at Time 2 from Adolescent Report

78

Figure 5: Mean Problem Behavior Total at Time 2 from Parent Report

79

Figure 6: Mean Problem Behavior Total at Time 2 from Adolescent Report

80

Figure 7: SSIS Scores at Time 1 and Time 2 for Parent and Adolescent Report

84

Figure 8: Difficulty Likert Scores at Time 1 and Time 2 for Parent and Adolescent
Report

84

Figure 9: Hyperactivity/Inattention Scores at Time 1 and Time 2 for Parent and
Adolescent Report

86

Figure 10: Bullying Scores at Time 1 and Time 2 for Parent and Adolescent Report

86

Figure 11: Internalizing Symptoms Scores at Time 1 and Time 2 for Parent and
Adolescent Report

87

Figure 12: Externalizing Symptoms Scores at Time 1 and Time 2 for Parent and
Adolescent Report

87

Figure 13: Problem Behavior Total Scores at Time 1 and Time 2 for Parent and
Adolescent Report

88

xii

KEY TO ABBREVIATIONS

AD: Asperger's Disorder
AS: Asperger's Syndrome
ASD: Autism Spectrum Disorder
ASSP: Autism Social Skills Profile
DSM: Diagnostic and Statistical Manual of Mental Disorders
GPA: Grade Point Average
HFA: High Functioning Autism
IAN: Interactive Autism Network
IDEA: Individuals with Disabilities Education Act
IMI: Intrinsic Motivation Inventory
MSPSS: Multidimensional Scale of Perceived Social Support
PDD: Pervasive Developmental Disorder
PDD-NOS: Pervasive Developmental Disorder-Not Otherwise Specified
RADS: Reynolds Adolescent Depression Scale
RCMAS: Revised Children’s Manifest Anxiety Scale
SSIS: Social Skills Improvement System

xiii

CHAPTER 1
Introduction
In the United States today, approximately one in 88 children is diagnosed with an autism
spectrum disorder (ASD; Center for Disease Control, 2012). In 2010, the Center for Disease
Control reported, “Assuming the prevalence rate has been constant over the past two decades, we
can estimate that about 730,000 individuals between the ages of 0 to 21 have an ASD.” As the
number of autism diagnoses has increased, so has the number of children receiving services
under the Autism classification in schools. The Individuals with Disabilities Education Act
(IDEA) Data Accountability Center reported that 93,660 children ages 3-21 were receiving
special education services under the Autism eligibility classification in 2000, and this number
climbed to 418,987 by 2010 (Data Accountability Center, 2011). As the diagnosis of ASD
becomes increasingly prevalent, there is a critical need to understand more about this disorder
and how to provide effective intervention for individuals who have ASD. ASD is a lifelong
disorder and there is no cure, so the challenges that individuals with ASD face often continue
throughout the lifespan. Therefore, it is important to provide developmentally appropriate
services and supports that enhance the quality of life for individuals with ASD.
In discussing the problems and challenges facing individuals with ASD, it is important to
clearly define the group of individuals referred to in this dissertation. According to the fourth
edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (American
Psychiatric Association, 2000), ASDs are classified as “Pervasive Developmental Disorders” and
consist of five different disorders: Autistic Disorder, Asperger's Disorder, Rett’s Disorder,
Childhood Disintegrative Disorder, and Pervasive Developmental Disorder-Not Otherwise
Specified (PDD-NOS). The DSM-IV diagnostic criteria (see Appendix C) state that children

1

with different disorders on the spectrum have different symptoms and therefore suggest that
these children may have different needs. For example, whereas children with Autistic Disorder
may have a general cognitive delay and delayed onset of language, children with Asperger’s
Disorder must not have had a clinically significant delay in language or in cognitive
development. There are however, two common diagnostic criteria across the autism spectrum:
1) a significant impairment in social interaction and 2) restricted, repetitive, and stereotyped
patterns of behavior, interests, and activities (American Psychiatric Association, 2000). More
specifically, children across the spectrum experience many difficulties with pragmatic language,
non-verbal cues, and interpersonal communication.
The recent publication of the fifth edition of the DSM (DSM-V; American Psychiatric
Association, 2013) substantially revised the way in which ASD is defined and diagnosed. The
DSM-V gathers Autistic Disorder, Asperger’s Disorder, Childhood Disintegrative Disorder, and
PDD-NOS into one category labeled, “Autism Spectrum Disorder”, and Rett’s Disorder becomes
its own category. The debate surrounding this revision is extensive and reflects the general
disagreement in the field regarding differences in diagnostic labels. Some researchers argue that
Asperger’s Disorder and Autistic Disorder are qualitatively different (see e.g., Neihart, 2000) and
others argue that Asperger’s Disorder and High Functioning Autism (HFA) are extremely similar
and can be studied as one (see e.g., Macinstosh & Dissanayake, 2006). Research has shown that
by adolescence those with HFA and Asperger’s Disorder have similar behavioral manifestations
despite their differences in early language development (Gilchrist et al., 2001). In schools, all
children with these disorders who are determined eligible for special education services are given
an Autism classification according to IDEA.

2

With such debate and disagreement surrounding the diagnostic categories, the labels become less
informative and a focus on individual needs becomes increasingly important in order to
effectively serve this population. Recently, the director of the National Institute of Mental
Health announced that it will be “re-orienting its research away from DSM categories” (Insel,
2013). There has been considerable debate about the validity of the recent revision of the DSM-5
and as a result, the director of the National Institute of Mental Health issued the following
statement: “Going forward, we will be supporting research projects that look across current
categories – or sub-divide current categories – to begin to develop a better system” (Insel, 2013).
Due to the shared characteristics of individuals across the autism spectrum, the IDEA
classification, the published DSM-5 revisions, and the recent National Institute of Mental Health
announcement, this dissertation identifies adolescents with any of the DSM-IV Pervasive
Developmental Disorders as “adolescents with ASD” and makes reference to this group as a
whole.
Challenges for Adolescents with ASD
Many adolescents with ASD have average to above average cognitive ability and are
therefore included in general education classrooms (Barnhill, Hagiwara, Smith Myles, &
Simpson, 2000). Although many have the skills to succeed academically, most adolescents with
ASD struggle to form positive relationships within the general education classroom and can
suffer socially and emotionally. In this setting, individuals with ASD face many difficulties with
pragmatic language, non-verbal cues, and interpersonal communication, all of which make
positive peer interaction very difficult. Common challenges include making eye contact,
accurately interpreting facial expressions and body language, staying on topic, appropriately
entering/exiting conversations, making jokes, understanding sarcasm, and differentiating

3

between language used with peers and with adults (Rao, Beidel, & Murray, 2008). Without the
requisite social skills, it becomes difficult for students to work successfully with others, adhere to
classroom behavioral expectations, and request help from others. In many high schools, students
have multiple teachers for all of their classes, requiring even greater knowledge of how to behave
and how to communicate in different contexts. Therefore, it is possible that these social
challenges could contribute to low academic achievement and poor mental health outcomes since
these students may have great difficulty interacting with others in a classroom setting.
For those adolescents with ASD who struggle socially and emotionally, treatment is
available in many different forms, such as school-based programs, intensive summer programs,
psychotherapy, or computer-based programs. These services may target various skill deficits
depending on the needs of the individual or the needs of a larger group. Many different
programs have demonstrated some success, but in many cases, it is difficult to attribute cause to
the intervention rather than other external factors, such as increased opportunities for peer
interaction, participant buy-in, and parent or teacher bias. Additionally, many programs take a
multimodal approach and include numerous intervention components. When social stories,
video modeling, parent involvement, and external reinforcement, for example, are all part of an
intervention, it becomes difficult to identify which component is responsible for change (BernadRipoll, 2007). There are few randomized-controlled trials that clearly demonstrate the
effectiveness of particular social skill treatments for adolescents with ASD and contribute to our
knowledge of evidence-based interventions (Solomon, Goodlin-Jones, & Anders, 2004).
One of the social skills intervention programs that is currently showing strong evidence
for effectiveness is the UCLA Peers Program (Laugeson & Frankel, 2010). Multiple randomized
controlled trials have indicated significant gains for those receiving treatment and a recent study

4

indicated positive maintenance effects as well (Laugeson, Frankel, Mogil, & Dillon, 2009;
Laugeson, Frankel, Gantman, Dillon, & Mogil, 2012). It is important to note, however, that
these studies took place outside of a school setting and required significant parent involvement.
Families face numerous barriers to accessing services however (Thomas, Ellis, McLaurin,
Daniels, & Morrissey, 2007), and therefore schools may be a more appropriate context for
providing these interventions to all deserving adolescents. However, there are also many
challenges to implementing evidence-based interventions in schools (Kratochwill & Shernoff,
2004) and therefore, many adolescents who could benefit from social skills interventions may
not receive them. In sum, many interventions for students with ASD have proven efficacious in
the research literature, and some have evidence for effectiveness, but few students may
ultimately receive these interventions, likely due to the numerous barriers to intervention
implementation.
Historically, there have been sparse data on the social skill challenges and subjective
experience of adolescents with ASD, presumably because much of the focus in empirical studies
has been on early childhood. Although early childhood research is valuable, research on
development beyond this period has been somewhat neglected (Schall & McDonough, 2010).
The skills taught to children with ASD in early intervention or elementary programs have not
been found to be sufficient to help individuals engage positively with peers in later years (Schall
& McDonough, 2010). Until recently, there were few studies that address the social skills that
become more relevant for adolescents or adults with ASD as developmental changes occur.
Some researchers (e.g., Elizabeth Laugeson, Fred Frankel, and Janine Stichter) have begun
addressing this gap and have published important research literature that focuses on this group of

5

older students. However, there continues to be a need for increased empirical research on the
social skills challenges and social experiences of adolescents with ASD.
Adolescence is a unique time of substantial growth and development for all students and
at times, it can be difficult and overwhelming to manage the many physical, cognitive, and
emotional changes. For adolescents with ASD, this period can be particularly challenging as they
become increasingly aware of their differences from neurotypically developing individuals
(Barnhill, Hagiwara, Smith Myles, Simpson, Brick et al., 2000). Though many adolescents with
ASD continue to show behavioral deficits in adolescence, many also show some improvement in
communication skills and increased social interest (Schall & McDonough, 2010). This does not
mean increased social success however, and students with ASD may become aware of their
rejection from peer groups and their lack of fulfilling friendships (Attwood, 1998). Data from the
National Longitudinal Transition Study-2 suggests that adolescents with ASD have less
confidence in their ability to make friends than adolescents in any other disability category
(Wagner, Newman, Cameto, Levine, & Marder, 2007). Adolescents with ASD are “significantly
more likely never to see friends (43.3%), never get called by friends (54.4%), or never be invited
to activities (50.4%)” compared to those with mental retardation, speech and language
impairment, or learning disabilities (Shattuck, Orsmond, Wagner, & Cooper, 2011, p. 5).
Finally, adolescents with ASD are less likely to participate in social activities with friends or
participate in community activities with peers than those in other disability categories (Shattuck
et al., 2011).
The social difficulties that emerge in adolescence often affect overall mental health for
those with ASD, as many go on to develop anxiety and depression (Church, Alisanski, &
Amanullah, 2000). Despite the fact that many adolescents with ASD suffer from social and

6

emotional challenges, only 22 percent receive mental health counseling or social work services
in school (Newman, 2007). In many schools, mental health services are a “supplementary item
on a school’s agenda” and are not well integrated to provide positive supports for students
(Adelman & Taylor, 2000, page 171). In middle schools and high schools in particular, the main
instructional focus often shifts from educating the whole child to focusing on core academics
despite the many developmental changes that are occurring for students. As a result, adolescents
with ASD who perform well academically may not receive appropriate services during the
school day to prevent or treat mental health problems. It is also possible that adolescents with
ASD who have negative social and emotional experiences may suffer academically during
adolescence. Creativity and flexibility are needed to serve these individuals, the majority of who
attend public schools (Newman, 2007).
The Internet may provide a mode of service delivery that is meaningful, appropriate, and
satisfying for students with ASD. In an online survey of 80 adolescents, 74% of respondents
reported that they prefer to communicate with others online (Hayter [Kuehnel], Oka, Tran,
Kloss, 2012). Additionally, some research suggests, “the use of computers likely creates an
environment for learning that appeals to individuals with ASD as less threatening” (Sansosti,
Powell-Smith, & Cowan, 2010). Finally, the Internet may provide a mode of service delivery
that is more cost-friendly, creating opportunities to serve more individuals. The Center for
Disease Control (2008) reported the economic costs associated with ASD are approximately 35
billion dollars per year, but these high costs have not led to appropriate services for all
individuals, and particularly adolescents, with ASD. Using the Internet to deliver interventions
to adolescents with ASD could potentially be a socially acceptable and cost-effective method.

7

Purpose of the Study
The primary purposes of this study were 1) to examine if an online social skills
intervention can improve perceptions of social skills or mental health problems (i.e. anxiety and
depression), 2) to determine how social functioning is related to mental health problems and
academic achievement, 3) and to identify differences in parent and adolescent perspectives.
Adelman and Taylor (2000) stated, “If schools are to work effectively on mental health
and psychosocial concerns, greater efforts must be made to develop comprehensive,
multifaceted, and integrated intervention approaches” (p. 171). It is important to focus on the
context in which individuals live and work and provide intervention in a way that best
accommodates diversity (Adelman & Taylor, 2000). Understanding the contexts in which
particular interventions and treatments are most effective is critical for providing the most
appropriate services to individuals. As adolescent Internet use increases exponentially (Gross,
2004), it may be helpful for researchers to consider online platforms as appropriate
developmental contexts for providing intervention and support. This study examined the Internet
as a novel context, and perhaps a more relevant and meaningful context, for the delivery of social
skills intervention services.
The ultimate goal of most social skills interventions is to increase an individual’s social
competence. There is an assumption inherent in social skills research that increasing social
competence is important because increased social competence is related to positive outcomes.
Research in developmental psychology has shown that stronger social skills are related to
increased social emotional functioning and higher academic achievement for typically
developing individuals (see e.g., Durlak et al., 2011; Tolan & Dodge, 2005; Wentzel, 1991), but

8

it is not clear whether this relationship is similar for adolescents with ASD. This study examined
this relationship specifically for adolescents with ASD.
Additionally, there is very little empirical data obtained from individuals with ASD,
themselves. The majority of studies rely solely on teacher, parent, or observer report to document
behavioral changes. If the ultimate goal is to create a more positive social and emotional
experience for the individual with ASD, then it seems vital to understand that experience
directly, rather than relying solely on the perceptions of others. This is especially true for
adolescents as they become increasingly aware of their disability and how their social skills
compare to others. Unlike most other social skills intervention studies, this study investigated the
perspectives of adolescents, in addition to parents, to assess similarities and differences in
reporting.
The Internet has become a common way of connecting people with common interests and
providing opportunities for individuals with ASD to interact online. WrongPlanet
(http://www.wrongplanet.net/) is an online community for individuals with ASD and other
neurological differences founded by Alex Plank, a young man with ASD. The discussion forums
on this website, and other Internet sources, provide a unique opportunity to hear the voices of
those with ASD and gain an insider’s perspective on the experience of having ASD. Plank has
become a major voice for the ASD community and has been featured in many news articles,
television programs, and online broadcasts. One of Plank’s most publicized and controversial
ideas is that ASDs should not be cured and that searching for a cure can be “damaging to people
who feel it’s part of their identity” (Deardorff, 2008). He does argue, however, that those with
ASD can improve and can learn, and that supports should be provided to help them do so. Plank
states, “For instance, spending hours trying to help your autistic child learn important life skills

9

is a good thing. Children with autism need to learn what's socially acceptable and they frequently
have a much harder time learning a lot of the things that neurotypical children already
understand” (Deardorff, 2008).
Although Plank is just one of many adolescents and young adults with ASD, his
perspective is valuable in that it highlights the lack of consensus among those with ASD, parents,
teachers, researchers, advocates, and other organizations. For example, when Barnhill et al.
(2000a) explored problem and adaptive behaviors in adolescents with Asperger’s from the
perspective of parents, teachers, and adolescents using the Behavior Assessment System for
Children, they found that results differed between parents, teachers, and adolescents. Parents
noted more severe behaviors than teachers, and adolescents saw themselves in the Average range
on the majority of subscales. The authors highlighted the disagreement among raters and called
for more research to investigate differences in perspectives. Additionally, they note that their
sample included both children and adolescents with Asperger’s and their results did not highlight
differences between the two, although adolescents with Asperger’s often become more aware of
their problems and differences (Barnhill et al., 2000b). This suggests that more research needs to
investigate the perceptions of those with ASD. The authors conclude, “Furthermore, an
understanding of these perceptions over various developmental ages would be valuable for both
professionals and parents. Such information could be used to support the development of
appropriate interventions for individuals with Asperger Syndrome, their families, and the
professionals who work with them” (Barnhill et al., 2000b, p. 164). This study examined both
parent and child perspectives on social skill development and social emotional functioning to
gain a more comprehensive understanding of the adolescent experience. Additionally, this study

10

investigated the utility of an online intervention approach from the perspective of the participants
to inform future intervention efforts.
Theoretical Framework
Given the various domains that this dissertation spans (i.e., autism spectrum disorder,
adolescent development, education, mental health, and technology), multiple theoretical
perspectives were influential in the development of the study. First and foremost, evidencebased practice is an important framework to consider for ASD intervention. Given the large
number of individuals with ASD and the diversity within that group, it is important to consider if
an intervention is truly effective and under what circumstances an intervention is effective. One
intervention strategy that has proven to be effective for teaching skills to individuals with ASD is
video modeling. Social learning theories provide theoretical support for video modeling as an
intervention strategy, while also providing a rationale for considering individual’s history and
experiences. Prior research provides support for video modeling as an evidence-based practice
for teaching those with ASD. Additionally, social learning theory addresses motivation to learn
as an important factor in observational learning, which led to careful consideration of key
intervention factors during intervention development (see Social Skills Intervention
Development below). Finally, the National Professional Development Center on Autism
Spectrum Disorders argues that an ecological systems approach is essential for promoting
effective practices for individuals with ASD. Therefore, in this study, Bronfenbrenner and
Ceci’s (1994) bioecological model provided a foundation for considering the multi-level
influences on adolescent’s development and the impact of having ASD in multiple contexts.
More specifically, adolescents’ development in context was considered as study measures were
selected in order to more broadly investigate the impact of social skills deficits across multiple

11

domains. Finally, sociocultural theories influenced the decision to gather data from multiple
perspectives. The emphasis in sociocultural theories on examining participant perspectives
aligns well with social learning theory’s focus on individuals’ experiences and with the
bioecological model’s acknowledgement of personal characteristics. Therefore, it was deemed
important to gather data from both adolescents and their parents.
Evidence-based intervention. With the current emphasis on evidence-based
interventions in education, it is important to examine the effectiveness of intervention options
available to those with ASD. There is a wealth of data on the effectiveness of early behavioral
intervention programs for children with autism (e.g., applied behavior analysis), but there is less
evidence for social skills interventions at later ages in development. In fact, Bellini, Peters,
Benner, and Hopf’s (2007) meta-analysis indicated that school-based social skills interventions
are only minimally effective for students with ASD. As the diagnosis of ASD becomes
increasingly prevalent, however, there is a critical need to understand how to provide effective
intervention for adolescents who have ASD. Knowing the context and population for which an
intervention is most effective can be a key factor in choosing whether or not to implement it.
High-functioning adolescents with ASD (i.e. those with average or above average cognitive
ability) comprise one subgroup that needs more critical attention. As stated previously,
adolescents with ASD who participate in the general education setting face numerous social and
emotional challenges but often do not receive school-based services. When services are offered
to adolescents with ASD, the selection of evidence-based interventions is hindered by the limited
number of methodologically sound studies that examine intervention effectiveness for
adolescents with ASD. Therefore, an evidence-based approach to social skills intervention

12

research will provide important information regarding intervention selection for adolescents with
ASD.
Social learning theories. In line with the evidence-based intervention approach
described above, the National Standards Project has identified modeling as an established
treatment, and the National Professional Development Center on ASD has identified video
modeling, in particular, as an evidence-based practice (National Autism Center, 2009; National
Professional Development Center, 2011). The theoretical support for video modeling strategies
comes from Bandura’s (1971) theory of observational learning, which was based on the premise
that individuals can learn through the process of observing others. Imitation of observed
behaviors, or performance, is most likely to occur when the behavior is presented in the proper
context and/or when reinforcement is provided. According to social learning theory,
observational learning can be used to explain the acquisition of many different behaviors, and
many successful interventions for children and adolescents with ASD are grounded in social
learning theory.
Video-modeling, in particular, is a strategy supported by social learning theory because it
provides structured opportunities for those with ASD to observe particular skills or behaviors
which they may need to learn. Video modeling is an instructional strategy where individuals
view themselves or others engaging in a particular behavior or task in a video. Bellini &
Akullian’s (2007) meta-analysis of 23 single-subject design studies using video modeling with
ASD populations suggested that skills acquired from video modeling are maintained over time
and generalized across settings. The authors concluded that video modeling could be considered
an evidence-based strategy for teaching social skills. It is important to continue assessing the
effectiveness of this strategy for adolescents with ASD however, as only three of the 23 studies

13

reviewed had adolescent participants. Additionally, all of the studies were conducted in home,
school, clinical, and community settings, but none were delivered online. Though the metaanalysis was methodologically sound and it seemed reasonable for the authors to conclude that
this strategy can be considered evidence-based, it is important to consider for whom and under
what circumstances this intervention is effective. Therefore, with strong theoretical and
empirical support, video modeling was chosen as an appropriate instructional strategy for the
online social skills intervention in this study.
Like all theoretical approaches, social learning theory has both strengths and limitations.
One strength of this approach is that constructs are typically easy to measure, as the behaviors of
interest are observable. Due to the specificity and measurability of such behaviors, hypotheses
are relatively easy to test. Social learning theory helps to explain the acquisition of new
behaviors and has strong practical applications, especially in reducing negative behaviors. The
limitations of social learning theory are also important however. First, the theory does not
address the issue of developmental constraint and suggests that observational learning functions
the same for individuals in various developmental stages and ages. Additionally, the theory does
not fully explain individual differences or the role of the individual in what he or she learns. In
this particular study, a primary limitation is that individuals’ behaviors will not be directly
observed and self-report and parent-report will be utilized as outcome measures (see Methods
section below).
Contextual development. The evidence-based intervention framework and social
learning theory both highlight the importance of development in context. When considering
whether or not an intervention is effective, it is important to ask the questions: For whom?
According to whom? Under what circumstances? (Ingraham & Oka, 2006). When considering

14

development from a social learning perspective, it is important to consider past influences,
current context, and possible systems of reinforcement and punishment. In general, both
approaches force one to acknowledge that learning and development do not happen in isolation
and that multiple factors are continuously at play. One way to consider the context of
development is through the bioecological model (Bronfenbrenner & Ceci, 1994). This model
calls for a consideration of the interaction between individuals and their surrounding
environment. In this study, the influence of the bioecological model is seen in the selection of
measures to examine potential moderating factors of influence on intervention effectiveness.
Another way to consider the contextual influences on development is through sociocultural
theories. Sociocultural theories (Daniels, 2005; Daniels, Cole, & Wertsch, 2007; Vygotsky,
1978) call for recognizing social, cultural, institutional, and historical influences. Specifically,
these approaches assume that meaning is socially constructed and to the extent that social
impairments are inherently a part of ASD, it is likely that individuals living with ASD experience
adolescence in a different way than neurotypical individuals. Sociocultural perspectives stress
the importance of exploring phenomenon from the perspective of the participants, and the voice
of those with ASD is one that is less often included in empirical research. This dissertation
addresses this gap by focusing on the voices of participants in order to understand the adolescent
experience from the perspective of those with ASD, while also valuing the perspective of their
parents. This allows for multiple interpretations of the data based on reporter.
Overall, an evidence-based intervention framework guided the study design, whereas
social learning theories guided the selection of an instructional strategy, and theories about
individuals’ development in context (i.e., bioecological theory and sociocultural theory) guided
the selection of study measures.

15

CHAPTER 2
Review of the Literature
This literature review will frame the goals and questions of this study within the context
of current ASD research. First, research on social skill deficits of children and youth with ASD
will be reviewed. Next, a review of the research on psychological adjustment outcomes for
children and youth with ASD will be presented. Then, a discussion of the intervention research
will follow, specifically highlighting the results and effectiveness of previous interventions for
social skill problems and psychological adjustment problems. The factors that affect
intervention success will be examined and finally, key factors for consideration in intervention
development will be reviewed. This section will conclude with the presentation of current
research questions and hypotheses.
Social Skills Problems of Individuals with ASD
Whereas the DSM-IV (American Psychiatric Association, 2000) articulates symptom
patterns of those diagnosed with Pervasive Developmental Disorders (see Appendix C),
numerous studies have attempted to more clearly describe the specific skill deficits of individuals
with ASD. Additionally, researchers have tried to identify skill deficits that distinguish those
with varying ASD diagnoses. Social skills challenges are typically present for all individuals
with ASD, regardless of the diagnosis. The results of such challenges often include socially and
emotionally inappropriate behavior, misunderstanding of social cues, inability to develop peer
relationships, and impairment in nonverbal communication, for example (Koning & MagillEvans, 2001). These challenges frequently occur throughout development given that the context
of social interaction changes. As individuals grow they must learn new skills and adapt.
Although children may be taught to understand social cues in early or middle childhood, they

16

may not understand social cues in adolescence without additional instruction. The studies
reviewed here primarily focus on participant samples that include children and youth with
Asperger’s Disorder (AD), high-functioning autism (HFA), and pervasive developmental
disorder-not otherwise specified (PDD-NOS), as these diagnostic labels most closely match the
participant sample for the current study.
Church, Alisanski, and Amanullah (2000) completed a retrospective chart review of forty
children diagnosed with AD between 1986 and 1998. This study was valuable in that it
highlighted children’s challenges across development. In elementary school, children had no
reciprocal friendships and only few superficial friendships. Children struggled to converse with
others and were often unable to take the perspective of others. Additionally, children’s socially
inappropriate interactions were thought to cause the alienation of classmates.
Data from 13 adolescents were analyzed to provide a description of social skills
challenges during the middle school and high school years. Half of the adolescents were able to
identify a best friend, but reported only speaking with this best friend only once or twice a
month. Friendships were primarily related to specific interests (e.g., playing Nintendo with a
friend). Rigid play was frequently described with adolescents having a difficult time breaking
from routine. Adolescents were also described as having difficulty staying on topic and
maintaining conversations. Adolescents had difficulty reacting to emotions in real life situations
and generally struggled with emotional expression. Finally, an inability to read social cues and
frequent peer rejection were also described. Generally, the challenges that exist in childhood
were seen to persist in adolescence.
To more specifically identify the social challenges of students with AD in comparison to
typically developing peers, Koning & Magill Evans (2001) compared 21 male adolescents with

17

AD with 21 matched controls on measures of social perception and broad social skills. Results
showed that adolescents with AD had lower social perception skills than matched controls, as
measured by the Child and Adolescent Social Perception measure (Magill-Evans, Koning,
Cameron-Sadava, & Manyk, 1995). There were significant (p = <.001) differences on the total
emotion score, as well as significant differences on the body, situation, and voice subscales that
comprise the total nonverbal cues score of the Child and Adolescent Social Perception Measure,
with typically developing adolescents performing better. The authors also found that the AD
group used facial cues more often than other cues (e.g., tone of voice). Broad social skills were
also found to be significantly different (p = <.001) between the AD group and controls, as
measured by parent-, teacher-, and self-report on the Social Skills Rating Scale (Gresham and
Elliot, 1990). On all three measures, the matched control group scored higher on the total social
skills variable. The assertion subscale of the Social Skills Rating Scale was also significantly
different (p = <.001), with controls scoring higher based on all three-reporters, whereas more
variation existed among reporters on the cooperation, self-control, and responsibility subscales.
A repeated measures ANOVA was used to test differences between raters and both group and
rater differences were significant (p = <.001) with no interaction effect. These results
demonstrate that those with AD struggle with social perception and various social skills,
although there appears to be some variation on specific domains of social competence and results
vary depending on the reporter.
Emotion recognition is another specific challenge that faces many individuals with ASD
and can impede their ability to successfully interact with others. Lindner and Rosen (2006)
examined the ability of youth with AD to decode emotion based on facial expressions, prosody,
and verbal language. Fourteen children and adolescents with AD were compared to 16 typically

18

developing children and adolescents on a measure of emotion perception (i.e., the Perception of
Emotion Test, Egan, 1989). According to Lindner and Rosen (2006), the Perception of Emotion
Test presents scenes in videotape format that are shown across multiple modalities using various
forms of verbal content, prosody, and facial expressions. Multivariate analyses indicated a
significant (p = <.01) difference of group by modality. Although there was not a significant
difference between groups on emotion perception with verbal content, the AD group scored
significantly lower when using facial expressions or prosody to make interpretations of emotion.
Overall, this study provides additional evidence that some individuals with ASD, particularly
those with an AD diagnosis, have challenges with forms of nonverbal communication.
Other studies have shown differences in the way individuals with ASD use verbal
information, however. Loveland, Pearson, Tunali-Kotoski, Ortegon, and Gibbs (2001)
investigated adolescents’ ability to recognize the appropriateness of social interactions. Nineteen
children and adolescents with autism were compared to 19 children and adolescents without
autism in their ability to judge the social appropriateness of 24 staged scenes. Half of the scenes
contained speech and half did not. After each scene, participants were asked if what they viewed
was okay or wrong. If participants believed the interaction was wrong, they were asked to
identify what was wrong and why it was wrong. The authors’ main finding was that, “Children
and adolescents with autism were less accurate in identifying examples of inappropriate social
behavior, but not appropriate behavior, than were those without autism, particularly where the
inappropriate social behavior was verbal” (Loveland et al., 2001, p. 372).
Other social communication challenges that individuals with ASD often face include
“choosing topics appropriate to the setting and the conversational partner, maintaining topics,
and indicating a switch to a new topic. They may have difficulty adjusting communication to the

19

needs of the person with whom they are speaking” (Krasny, Williams, Provencal, & Ozonoff,
2003, p. 107). Rubin and Lenin (2004) identified eight particular capacities for social
communication that those with ASD often struggle with: determining the attentional focus,
perspective, and intentions of others; perceiving emotional states and considering plausible
causal/actors; initiating and maintaining conversational exchanges; recognizing and repairing
breakdowns in communicative exchanges; understanding and using more sophisticated language
as a means to clarify intentions; understanding and using nonverbal cues; understanding and
adhering to social conventions and cultural norms for initiating, exchanging turns, and
terminating interactions; using language as a tool for guiding behavior, collaborating with others,
and emotional regulation.
Overall, the challenges described above are problematic for those with ASD and are also
long lasting. In a study of perceptions of communication skills, typically developing adolescents
rated perspective taking and vocal tone comprehension as the two most important
communication skills in positive peer relationships (Henry, Reed, & McAllister, 1995). If
adolescents with ASD lack the skills that typically developing peers believe are important, they
will likely have difficulty developing positive peer relationships. These challenges appear to
persist into adulthood and can result in feelings of isolation and longing for intimacy and social
connectedness (Muller, Schuler, & Yates, 2008). In describing specific challenges that
contributed to difficulties navigating the social world, adults with ASD described difficulties
initiating social interactions and also specifically noted, “participating in impromptu and/or
unstructured dialogue requiring improvised responses, understanding implicit as well as explicit
meanings of messages, drawing social and emotional inferences, and interpreting and using
gesture and tone of voice (e.g. sarcasm, irony or innuendo)” (Muller, Schuler, & Yates, 2008, p.

20

179). Ultimately, these challenges left many adults with ASD craving emotional intimacy,
wanting deeper friendships, and desiring romantic relationships.
Psychological Adjustment Problems for Individuals with ASD
Problems with psychological adjustment become most concerning as students with ASD
who have average to above average intellectual functioning enter adolescence. Adolescence is a
critical time for developing peer relationships and the impairments associated with ASD make
social interaction extremely difficult. The core impairments that are characteristic of ASD can
also lead to other psychological disorders, such as anxiety and depression (Tantam, 2000).
Many children with ASD diagnoses who do not have diagnosed anxiety disorders
experience more anxiety than typical children (Russell & Sofronoff, 2005). Sixty-five children
with AD were compared to children with clinical diagnoses of anxiety and typically developing
children on the Spence Children’s Anxiety Scale (Spence, 1998) and the Social Worries
Questionnaire (SWQ; Spence, 1995). According to both parent and child report, children with
AD have significantly higher (p < .01) overall levels of anxiety than typical children. Parents
also indicated that children with AD have significantly higher obsessive-compulsive symptoms
and physical injury fears (p < .05).
In another study of 29 adolescents with ASD and 30 nonclinical adolescents, Farrugia
and Hudson (2006) also found the ASD group to have significantly higher anxiety than the
control group (p < .05) on the Spence Children’s Anxiety Scale. Additionally, the ASD group
had significantly higher negative thoughts, behavior problems, and life interference than
adolescents with anxiety disorders. Another study, by Kuusikko et al. (2008) found that
generalized social anxiety in particular, is higher for adolescents with ASD than typically

21

developing adolescents, as measured by the Social Anxiety Scale for Children-Revised (La
Greca & Stone, 1993).
Recently, Mazurek and Kanne (2010) provided supporting evidence that adolescents with
ASD have higher levels of internalizing symptoms, as measured by the Child Behavior Checklist
(Achenbach & Rescorla, 2001) than those without ASD. Additionally, they found that those
with higher ASD symptoms have lower levels of anxiety and depression, contrary to their
hypothesis. It is possible that those with fewer ASD symptoms are more aware of their social
difficulties and lack of friendships, and are therefore more likely to experience anxiety and
depression. These results are particularly interesting when paired with Barnhill’s (2001)
previous findings that the higher an adolescent’s IQ, the less likely he/she was to attribute social
success to task and chance. In Barnhill’s (2001) study it was also found that adolescents with
more depressive symptoms were more likely to attribute social failure to ability and to the sum of
their ability and effort.
Social comparison and perceptions of group membership may also be related to
depressive symptoms for adolescents with ASD. Hedley and Young (2006) investigated the
relationships between social comparison processes and depression in adolescents with AD.
Participants included 36 adolescents who completed the Social Comparison Scale (Allan &
Gilbert, 1995; adapted by Dagnan & Sandhu, 1999) and the Children’s Depression Inventory
(Kovacs, 1992). Though cautious interpretation is warranted due to methodological limitations,
the study suggests that adolescents who perceive themselves to be much different from others are
more likely to display depressive symptoms (r = 0.52, p = 0.001).
Adolescents with ASD also experience significantly (p < .005) more loneliness that those
without ASD based on self-report on the UCLA Loneliness Scale (Lasgaard, Nielsen, Eriksen, &

22

Goossens, 2010). In the same study, loneliness was also found to negatively correlate with
perceived social support from classmates (r = -.52, p = < .005), parents (r = -.43, p = < .01), and
a close friend (r = -.40, p = < .05) for adolescent boys with ASD. This suggests that social
support may protect against loneliness for this population (Lasgaard et al., 2010).
Numerous studies have demonstrated that feeling part of one’s school community, or
having a sense of school membership, is positively correlated with academic achievement for
typically developing individuals (see e.g., Goodenow, 1993). It is not known if this correlation
exists for those with ASD as well. This relationship is important to investigate so that
practitioners can monitor whether or not students with ASD feel like they belong in their school
communities and maximize opportunities for academic achievement.
Though parents are typically asked to report on their child’s social and emotional
functioning through parent report forms of standardized rating scales, it is ethically and
scientifically important to gain adolescents’ perspectives on their social and emotional
functioning. The findings reported above have demonstrated some discrepancies in perceptions
between parents and adolescents. High functioning adolescents with ASD are capable of
providing their perspective and should be included in both clinical and empirical assessment.
Preliminary evidence has suggested that adolescents’ reports of quality of life are valid and
reliable and are also negatively correlated with measures of anxiety (Shipman, Sheldrick, &
Perrin, 2011). This suggests that assessing adolescents’ perspectives on life experiences, as well
as their thoughts and emotions, is important for promoting positive mental health outcomes.
Success of Previous Interventions
As the prevalence of ASD has increased, so too have the number and type of programs
available to individuals with ASD diagnoses (Autism Speaks, 2010). Social skills training

23

programs have been a popular form of intervention to address the social skills challenges that
individuals with ASD often face. Various methods of teaching social skills include role-playing,
social stories, modeling, computer games, video games, and direct instruction. In addition, the
method of instruction often varies between programs and instruction may be delivered
individually, in small groups, or in large groups. Finally, social skills curricula may vary both
within and between programs depending on the client’s skill deficits. Many published articles
have focused on the success of social skills programs for individuals with ASD. To review all of
them would be excessive for this dissertation, and therefore a sample of articles have been
chosen to highlight the range of intervention programs and strategies that can be found in the
research literature. General trends among the published studies will be discussed, though it is
important to note that this literature review is by no means comprehensive.
Group interventions. Group interventions are often manualized programs where all
children in a group are taught the same curriculum and given the same instruction. These
programs are meant to offer all children the same opportunities for learning by giving them tools
to help each other learn. These interventions are based on the idea that children learn best in
context and that social skills should be taught in social situations. A common presupposition of
group social skills interventions is that by teaching specific skills and including all members of
the learning community (parents, teachers, children), children have the opportunity to succeed
both socially and academically.
Barnhill, Cook, Tebbenkamp, & Smith Myles (2002) implemented an 8 week social skills
group for 8 participants (7 male, 1 female) with ASD, consisting of direct instruction, roleplaying, modeling, and reinforcement through feedback, and focused primarily on teaching
paralanguage and facial expressions. At the end of 8 weeks, results suggested no significant

24

differences on the Diagnostic Assessment of Nonverbal Accuracy 2 scores between pre- and
post-tests using the Wilcoxin signed-rank test. Despite the lack of significant findings however,
the authors claim that visual examination of the data reveals a slight increase in scores from pretest to post-test. Parental feedback was also strong: 100% of parents wanted their children to
continue in the group, 75% believed their child had developed friendships, and 57% believed
their child would call a friend from the group on the phone. The one female participant reported
feeling isolated in a group of all boys, but the remaining 7 male participants all reported that they
had made friends in the group.
Bauminger (2002) described an intervention program aimed at increasing socialemotional understanding and social interaction among adolescents with ASD. A 7-month,
multidimensional program was implemented for 15 children between the ages of 8 and 17 and
focused on social cognitive capabilities including social problem solving, emotional
understanding, and social interaction abilities. Each participant worked on the intervention
curriculum for three hours per week with his or her teacher in class. Additionally, participants
met with assigned peers twice per week for peer support, and parents were contacted by teachers
on a weekly basis so they could provide support and motivation at home. Social problem solving
was measured by an adapted form of the Problem Solving Measure (Lochman & Lampron,
1986), emotional understanding was measured by the Emotion Inventory (Seidner, Stipek, &
Feshbach, 1988), and social interaction abilities were measured by researcher observations and
the Social Skills Rating Scale—Teacher Version (Gresham & Elliot, 1990). Children showed
improvements in social problem solving, emotional understanding, and social interaction abilities
across the intervention period and the authors concluded that intervention appeared to be
effective, but also acknowledged that it was not possible to make firm conclusions about

25

causation without a control group. The authors also acknowledged that the design of the study
did not enable them to determine which element of multimodal intervention was responsible for
participant change. Other factors such as adolescent maturation or the attention and support of a
researcher or teacher could be related to participant change.
Lopata, Tomeer, Volker, Nida, and Lee (2008) investigated the benefit of another group
intervention program designed specifically for children with ASD. Their study closely explored
an intensive summer program for teaching social skills to 54 children (ages 6 to 13) with HFA
and AD. In particular, Lopata and colleagues were interested in evaluating the effect of the type
of feedback children received during intervention. Children were assigned randomly to either a
response-cost feedback group or a feedback group with no behavioral categories or
contingencies, and both had manualized instruction from undergraduate and graduate students in
psychology and education. The program, Skillstreaming, placed a strong emphasis on teaching,
modeling, role-playing, performance feedback, and transfer of learning. Overall program results
suggested a significant increase in social skills from pretest to posttest, as measured by the Social
Skills index of the Behavior Assessment System for Children, according to both parent (p =
.048) and staff report (p = .019). The authors did not find significant differences between the two
treatment conditions. Results showed that high rates of performance feedback and reinforcement
are significant in promoting effective social skills. The authors also concluded that there is
support for the idea that social skills interventions should break behavior down into components
and teach skills in part-to-whole sequence while providing feedback. It is important to note that
the positive results came from parents and staff and not from the children themselves.
In another study, parent and teacher reports also confirmed positive results using a
different, and rather unique, technique to teach social skills in group settings: group

26

psychotherapy. Tyminski and Moore (2008) investigated the effect of long-term group
psychotherapy on the improvement of social skills for 39 children (ages 5 to 16) diagnosed with
HFA, AD, or PDD-NOS. The duration of treatment varied depending on the client (i.e. four
months to three years) and follow-up assessments were completed at 14 months on average. The
children's social development was assessed at home and school, both before and after treatment.
Significant improvements were found in both settings and neither was correlated with
demographic variables or cognitive development, which according to the authors, implies that
the change is due to the intervention. The authors conclude that group psychotherapy is a good
way to understand the social skills development and treatment in the many children diagnosed
with ASDs.
Another novel study was conducted by Licciardello, Harchik, and Luiselli (2008).
Intervention-facilitated social initiation was the focus of the study, as opposed to the more
commonly studied social response. Baseline measures were taken and then a social skills
intervention, which included pre-teaching, prompting, praise, and reward, was implemented for
four students between the ages of 6 and 8 with ASDs. Observations were recorded by the
research team and based on their data, the authors concluded that the number of social initiations
and responses "meaningfully increased" throughout the intervention. Note that the results are
based on researcher report of behavior and therefore reflect the researchers’ assumptions about
socially appropriate and inappropriate behavior.
More recently, Stichter and colleagues (2010) published a study that examined the
effectiveness of a group-based social competence intervention for adolescents (aged 11-14) with
HFA and Asperger's. This particular study utilized cognitive behavioral techniques to improve
adolescents’ skills in perceiving and understanding emotions, theory of mind, and executive

27

functioning. Over twenty hours of intervention (i.e. one hour group, twice a week after school,
for ten weeks), 29 male adolescents made significant improvement in facial expression
recognition, theory of mind, and problem solving. Though these results were promising, the
authors suggested that this intervention needs further investigation through larger sample sizes
and an exploration of generalization through application in naturalistic settings. An initial
investigation of generalization revealed some social skill improvement and gains in executive
function, based on teacher report, and some improvement in facial recognition based on direct
observation (Schmidt, Stichter, Lierheimer, McGhee, & O’Connor, 2011).
To many, randomized controlled trials are considered the gold standard in research
practices (Chambless & Hollon, 1998). Beaumont & Sofronoff (2008) researched the effects of a
social skills intervention program called The Junior Detective Training Program by randomly
assigning 49 elementary-aged children with AD to an intervention or wait-list control condition.
The seven-week program had multiple components including computer games, small group
session, parent training, and teacher handouts. Through these various modalities, the goal of
intervention was to improve skills in emotion recognition, emotion regulation, and social
interaction. On a parent-report social skills measure (i.e., the Social Skills Questionnaire-Parent),
the intervention group made significant improvements from pre- to post-treatment (p < .001)
whereas the wait-list controls did not. In addition, the scores improved from the Clinically
Significant to Normal range, suggesting clinically meaningful gains. On measures of emotionrecognition, the treatment group did not make greater gains than wait-list controls. On the
measure of emotion regulation however, the treatment group improved more than controls and
maintained gains at 6-week follow up. Finally, there was not enough teacher-reported data to

28

determine generalization effects. The authors concluded that the intervention was effective in
improving social skills and emotional understanding.
Williams White, Keonig, and Scahill (2006) conducted a review of group-based social
skills training programs for school-aged youth with ASD. The studies in Williams White,
Keonig, and Scahill’s (2006) review varied in method and included pre-post, controlled, and
single-subject designs. Overall, the review identified several promising treatments, but also
highlighted numerous methodological weaknesses of the reviewed studies. Many studies
showed non-significant changes in social skills, differences in outcomes based on reporter, and
poorly defined social skills deficits.
In general, the individual studies reviewed here, as well as the results of a comprehensive
review (i.e., Williams White, Keonig, and Scahill, 2006) indicate that results vary when teaching
social skills in group contexts. Furthermore, there are various ways of teaching social skills: inschool universal programs, intensive targeted summer programs, after-school cognitive
behavioral groups, or targeted psychotherapy sessions, for example. With success across
programs, it is difficult to draw conclusions about what the critical causal mechanism is for
improvements in social skills for children and youth with ASD. Also, since different authors
define social competence in multiple ways, it is difficult to determine which social skills are
most important for developing positive peer relationships.
Individual interventions. Researchers have also investigated the effects of
individualized programs. Individual interventions more specifically address the unique
behaviors of individuals rather than generalizing curriculum to large groups of students. This is
generally thought to be an effective strategy with individuals who have ASD because there is
such great variation in individual behaviors.

29

For example, Frederickson, Simmonds, Evands, and Soulsby (2007) conducted a study
assessing the social and affective outcomes of inclusion. The authors compared three groups of
children: those previously attending a special school who had recently transitioned into an
inclusive classroom, those with special needs who had always been mainstreamed, and typically
developing students. A total of 397 children aged eight to eleven participated, 14 of who had
recently transitioned. Twelve of those 14 children had been diagnosed with ASD and were
receiving individualized support from an outreach program team. The children were compared
on a number of measures, such as social acceptance, belonging, and bullying. Students who
were transitioning into an inclusive classroom and had an extensive team to help with this
transition experienced positive emotional and affective outcomes and reported feeling like they
belonged, as measured by The Belonging Scale (adapted from the Psychological Sense of School
Membership; Goodenow, 1993). The special education students who had always been
mainstreamed and did not have this team experienced many more negative social and affective
outcomes. What this study shows is that tailoring an intervention to the needs of an individual or
buffering a new social experience with individual supports may produce more positive outcomes
(as the outreach team did in this study; Frederickson et al., 2007).
Sansosti and Powell-Smith (2008) similarly showed that individualized interventions
could have a positive effect. They investigated the effects of computer-presented social stories
and video models on the social communication skills of three boys (aged six to ten) with an ASD
diagnosis. Each social story and video model was tailored to the child’s specific skill deficits
and a multiple-baseline across-subjects design was used to evaluate change. The social stories
and video models were used in the classroom setting and observation data was collected in order
to track the improvement of skills over time. The authors observed the students in various

30

settings to document generalization effects and also performed follow up assessments to track
maintenance effects. Various modifications (e.g., teacher prompting) to the intervention were
made suggesting that students’ individual needs are crucial to developing a successful
intervention. Overall, the authors concluded that the intervention was effective, but noted that
because individual modifications were made the computer-presented stories and video models
may not have been the effective components of change. Ultimately, it is impossible to tell
whether the stories, computer, researcher, teacher, or attention the child received was the cause
of change.
Similar to group interventions, individual interventions leave researchers, practitioners,
and families with more questions than answers. What these studies have shown is that two other
tactics (i.e. outreach teams, and computer-presented social stories and video modeling) have
some positive effect on the increase of social skills in individuals with ASDs. It is unknown
from these two studies, however, why these programs change behaviors, how long the changes
last, and how the interventions can be generalized.
Numerous case studies exist in the research literature that demonstrate improvement in
social skill development for individuals with ASD. It would be beyond the scope of this review
to discuss all of them or attempt to synthesize the data. However, it is important to note that
emphasizing the needs of individuals and the context in which individuals develop may be an
important aspect of intervention implementation. Moreover, single-case research has
demonstrated great effectiveness of social skills interventions for individuals, particularly when
multiple-baseline and reversal designs are used (Wang, Parrila, & Cui, 2012). Wang, Parrila,
and Cui (2012) conducted a meta-analysis of 115 single-case studies with 343 participants and
found that overall, social skills interventions tend to be effective for those with ASD. They also

31

found, however, that length of intervention, quality of intervention, gender, and age of
participants, were not significant predictors of effectiveness. They did find however, that study
design was predictive of effectiveness and more specifically, that multiple baseline and reversal
study designs more frequently showed better outcomes. This suggests that the way in which one
measures outcomes is particularly important when drawing conclusions about intervention
effectiveness for this population.
Interventions with peers. Multiple studies have shown that including typically
developing children in intervention implementation can be beneficial for students with ASD.
Harper, Symon, and Frea (2007) designed a study to address how naturalistic techniques can help
improve social skills in children with ASD by testing the effects of a peer-mediated intervention
for children with ASD in an elementary school. More specifically, they observed two thirdgrade students and how they learned social skills from six peer volunteers. These volunteers had
been chosen to learn particular strategies and skills for working with individuals with ASD and
had been trained over many weeks to learn how to use cards and cues when interacting with their
peers with ASD. The dependent variables were specific to each individual and determined by
observation, teacher input, and individualized social goals. For example, one student’s
dependent variables were the number of turn-taking interactions and the number of attempts of
gaining peers’ attention. There was a large increase in prosocial behavior in the two students
with ASD and the study concluded that peer-mediated naturalistic social skills instruction can be
effective.
In another study, Knott, Williams, and Lewis (2007) also observed how peers could
influence social skills development, but studied family members. Contrasting Harper et al.’s
(2007) study of school peers, Knott and colleagues looked at how siblings could influence social

32

learning. Sibling relationships have been noted as a key point of study in examining how
children with ASD acquire social skills, but according to Knott et al. (2007), there has been very
little longitudinal research. In this study, siblings were not trained and no intervention was
implemented. Rather, naturalistic observation was used to identify patterns of interaction
between typically developing children and their developmentally disabled siblings (either
diagnosed with autism or Down Syndrome). The study looked at 16 sibling dyads (6 containing
a child with ASD and 10 containing a child with Down Syndrome) over 12 months. Changes in
the amount and rate of both prosocial and agonistic interaction were measured using behavioral
observation.
The results showed that prosocial initiations significantly (p = < .0001) increased in
children with ASD over a 12-month period, as well as the frequency of imitation (p = < .005).
The authors conclude siblings can play a special role in social and emotional development, and
more specifically that interventions grounded in existing interactions have the potential to more
naturally improve a child’s patterns of interaction. Similar to Harper et al.’s (2007) study above,
this study provides support for the idea that learning in the context of relationships is important.
Although this one study has limited generalizability, it informs the reader about the
values underlying peer-mediated interventions. Learning cannot be separated from context and
this perspective is very evident in this study. Additionally, there are many published single-case
studies that also examine peer-mediated interactions and show strong effects, which suggests
some generalizability to other participants and contexts. These few studies of peer-mediated
interventions lend support to the claim that there is little conclusive evidence on what specific
intervention is most effective. Whereas one study identified school peers who had been trained
to use cue cards as effective agents of change, another study identified playtime with siblings as

33

an effective method of learning. Both studies use different interventions, yet each author claims
support for the intervention. Taken together, these studies suggest that although many different
interventions may be effective, there may be some specific methods or strategies that are likely
to be effective across individuals.
Video modeling interventions. Since many students with ASD have visual processing
strengths, video-based interventions have become a more common form of social skills training.
This approach is discussed separately from the above group, individual, and peer interventions
because video modeling can be delivered in many different formats. The potential strengths of
video modeling interventions are abundant, but particularly include the opportunity for frequent
repetition in the presentation of material. Ayres and Langone (2005) write, “Consider the time
investment needed to have a model (whether a teacher or peer) repeat a behavior several times.
Repetition is important but what if the student trying to learn the skill requires more
opportunities to view the model than the model has time to perform?” (p. 183). A review of 15
interventions that include the use of video as an intervention tool to teach social skills and
functional communication found generally positive results across studies, and therefore did not
identify a specific component of video use that leads to improved outcomes (Ayres & Langone,
2005). One important question that remained at the conclusion of this study was which elements
were responsible for change? (e.g., Do self-as-model videos produce better outcomes than otheras-model?).
This question has been addressed in numerous studies however (see e.g., Sherer et al.,
2001), and ultimately, there are no significant differences between self and other video models.
Sherer et al. (2001) used a multiple baseline alternating treatments design with five children to
examine rate of skill acquisition with both self and other video models. They found that children

34

who learned from video modeling learned equally well from both conditions. In general, video
modeling is yet another intervention strategy, which may fall into any one of the above
categories of intervention. Video modeling may be used on an individual basis or in groups, and
may also utilize peers to model appropriate behaviors for individuals with ASD. It has shown
some promise in recent literature and is considered a promising evidence based practice
(Reichow & Volkmar, 2010).
Factors That Affect Intervention Success
Overall, many forms of social skills instruction appear to be efficacious, and a recent
review has indicated that there is “much empirical evidence supporting many different treatments
for the social deficits of individuals with autism” (Reichow & Volkmar, 2010, p.149). Evidencebased practices often revolve around manualized curricula, and although numerous manualized
curricula can be purchased to teach social skills to children and adolescents with ASD, each may
embrace different methods and modes of delivery, and none are necessarily appropriate and
effective for all individuals with ASD. A better approach may be to consider guidelines for best
practice that can be utilized within various intervention approaches.
Best practice guidelines for effective social skills training with individuals who have
ASD include four steps: 1) identifying/targeting specific social skills, 2) distinguishing between
skill and performance deficits, 3) direct and systematized instruction, and 4) monitoring student
performance (Sansosti, Powell-Smith, & Cowan, 2010). These are basic steps that fit within a
problem-solving model grounded in data based decision-making.
Identifying and targeting specific social skills is an important part of this process, as
much of the literature above demonstrates that individuals with ASD may have different needs
across the lifespan. Additionally, those with ASD may have varying perceptions of which skills

35

are most relevant and important in their social contexts. Distinguishing between skill and
performance deficits is also a logical step in order to determine the type of support or instruction
that will be most supportive for the individual. Choosing how to provide direct and systematized
instruction is perhaps a more challenging step in this process.
According to Sansosti, Powell-Smith, & Cowan (2010) there are six important steps to
systematic instruction: 1) provide rationale for the skill to be taught, 2) preview the skill to be
taught, the activities, and the time, 3) discussion, 4) modeling, 5) role-playing, and 6)
constructive evaluation of performance. The research highlighted above has demonstrated that
there are multiple ways to teach social skills effectively. However, there does seem to be some
evidence that strategies that embrace elements of discussion, modeling, and practice appear to be
most effective. Finally, the fourth suggested step of monitoring progress is important to
determine if intervention is in fact leading to the intended outcomes.
Key Factors for Online Intervention Development and Implementation
Based on the reviewed research on challenges for those with ASD and successful
intervention strategies, video modeling was selected as the intervention mechanism for the
current study. In comparison to studies on video modeling discussed previously in this review,
the goal of this study was to investigate video modeling delivered asynchronously over the
Internet. Therefore, a brief review of video modeling procedures, a rationale for online
intervention, and a discussion of the intervention structure for the current study are appropriate.
Video modeling intervention procedures. Video modeling interventions involve a
“procedure in which a learner is shown a videotape of a model performing a target behavior or
completing a desired task” (Sigafoos, O’Reilly, & de la Cruz, 2007, p. 1). Video modeling may
take place in a variety of contexts (e.g., school, home, therapy, etc.) and can be done with self-as-

36

model or other-as-model. Most often, “after viewing the entire videotape-from start to finish-the
learner is given the opportunity to perform the modeled behavior or complete the task that was
demonstrated in the video” (Sigafoos, O’Reilly, & de la Cruz, 2007, p. 1). Practice
opportunities, immediate feedback, and reinforcement are thought to be critical intervention
elements. Other researchers, Sansosti, Powell-Smith, and Cowan, have most clearly described
the necessary video modeling procedures with the following six steps:
1) Select a skill to be taught by the video model.
2) Break the target skill down into smaller, teachable units.
3) Write a script for the video model.
4) Train models to engage in the skill.
5) Record the model and edit the video clip.
6) Implement the video modeling intervention and monitor effectiveness (2010, p. 147).
Throughout the video modeling literature, it is apparent that opportunities for repeated viewing
of the target skill are also typical of video modeling interventions.
Using these video modeling procedures, many researchers have demonstrated the
effectiveness of video modeling, which is why it has come to be regarded as an evidence-based
practice (Bellini & Hopf, 2007). The utility of video modeling for computer based interventions
however, is less well established. Given the evidence that video modeling interventions are
efficacious, it seems appropriate to investigate more efficient ways of delivering video modeling
interventions to children and youth. If research can provide evidence that the computer-based
delivery of video modeling is an instructionally sound practice, this may provide an efficient and
effective way to make treatment widely and easily accessible.

37

Rationale for online interventions. When developing a computer-based or online social
skills intervention, it is important to consider evidence-based practices for social skills
intervention, the unique features of computers and the Internet, and research on adolescents’ use
of technology. Each of these considerations is important to ensure that the intervention structure
is most appropriate for the participants receiving the intervention.
Based on the existing social skills intervention literature, Krasny et al. (2003) outlined
several important elements for successful group intervention for individuals with ASD (see Table
1). They state, “These principles can be used to develop new curricula, modify existing curricula
to make them more accessible and relevant to people with autism, or evaluate programs to
determine their potential usefulness for people with ASD. These essential ingredients are not
mutually exclusive, nor are they an exhaustive list of all possible best practices in social skills
training” (Krasny et al., 2003, p. 110). Two of the central principles identified by Krasny and
colleagues (2003), were to “provide multiple and varied learning opportunities” and to “provide
opportunities for programmed generalization and ongoing practice” (p. 111). Computer based
interventions offer both of these opportunities.
Using computers to facilitate skill development for individuals with ASD has become a
promising strategy and has been investigated in numerous research studies. Intervention
strategies that engage technology (e.g., video-modeling, computer-presented Social Stories, etc.)
have become common in the research literature, but the majority focus on computer or
technology use in the context of a laboratory, classroom, or face-to-face group intervention (see
e.g., Silver & Oakes, 2001). For high functioning adolescents with ASD who participate in
general education classes, participate in after school activities, or hold part-time jobs, classroom
instruction or face-to-face intervention opportunities are limited. All of the intervention strategies

38

discussed above may simply not be appropriate for a particular group of students with ASD (i.e.
higher functioning adolescents who do not access special education supports in schools or
Table 1
Krasny et al.’s (2003) Essential Ingredients of Group Social Skills Curricula
Central principles
Make the abstract concrete

Implementation examples
Explicitly define skills/problems
Use ‘‘if-then’’ rules
Provide visual cues and prompts
Provide structure and predictability
Follow consistent weekly routine
Use consistent opening and closing formats
Provide schedules or lists of group activities
Engage in concrete tasks that ease transitions
Provide scaffolded language support
Group children by language level
Simplify language
Use visual supports
Provide language models and scripts
Provide multiple and varied
Offer multisensory opportunities
learning opportunities
Vary activities, materials, and teaching techniques
within and across sessions
Alternate size of work group
Include ‘‘other’’-focused activities
Always work in pairs or groups
Foster cooperation and partnership
Teach perspective-taking skills
Highlight peers’ preferences and interests
Foster self-awareness and self-esteem
Create positive, nurturing environment
Identify individual strengths and positive attributes
Examine ASD strengths and positive attributes
Provide constructive feedback
Select relevant goals
Prioritize skills that are most salient to ASD
Utilize socially relevant activities
Provide rationale for skill use
Individualize goals and skill development
Program in a sequential and
Simplify complex behaviors into specific skills
progressive manner
Teach skills sequentially
Integrate and practice mastered skills together
Provide opportunities for programmed
Use multiple and varied learning opportunities
generalization and ongoing practice
Provide generalization activities for home & school
Practice with varied people and settings
Offer social skills training in a school
______________________________________________________________________________

39

therapeutic supports outside of school). Providing opportunities for students with ASD to
participate in the least restrictive environment, while also providing supports and resources for
the challenges associated with ASD, presents a complex issue. One possible solution is to
provide intervention content and resources, as well as opportunities to interact with others, in an
online community or forum. Additionally, this mode of service delivery allows adolescents to
improve their skills in a more private way that avoids labeling or the stigma that may be
associated with school-based mental health services.
The Internet has become a popular mode of communication for those with ASD as it
eliminates many of the challenging nonverbal components of interpersonal interaction. It
appears that many adolescents with ASD prefer communicating on the Internet and therefore it
could be a successful method for delivering intervention (Burke, Kraut, & Williams, 2010).
Although the effectiveness of the Internet as a mode of delivery has not been examined yet,
computer-aided instruction was recently identified as an emerging practice in the National
Standards Project (National Autism Center, 2009) and was identified as an evidence-based
practice by the National Professional Development Center on ASD (National Professional
Development Center, 2011). Numerous articles in popular media have been tackling this issue
(see e.g., Blume, 1997) and the large numbers of adolescent members of online communities,
such as WrongPlanet.net, indicate a preference for computer-based communication. Multiple
explanations have been proposed for why those with ASD may prefer online communication.
Benford (2008) writes, “Online communication may compensate for difficulties with spoken
and/or face-to-face social interaction and the lack of offline social interaction which may result
from such barriers. The different synchronicity of Internet-based communication, the use of text
rather than speech, and the lack of nonverbal communication have been raised as possible factors

40

which may make online interaction easier, perhaps more comfortable, for people with HFA/AS.”
In Benford’s (2008) online survey of 138 adults with HFA/AS, respondents indicated their top
two reasons for using the Internet were to have contact with others who had a shared hobby or
interest and because they simply enjoyed this type of communication.
There are many assumed benefits to delivering social skills instruction online. In fact,
“Virtual communities can be seen as mental health and social support interventions” in and of
themselves (Eysenbach, Powell, Englesakis, Rizo, & Stern, 2004). Online communities provide
opportunities for individuals who are geographically distant to communicate in real time. People
with no prior acquaintance who share similar interests are able to connect in online communities
and this connection can often result in members providing one another with social support
(DiMaggio, Hargittai, Neuman, & Robinson, 2001). Numerous studies have been completed
examining the health benefits of online support networks and virtual communities and the results
of those studies are mixed (Eysenbach et al., 2004). There are few studies that explore the utility
of online communities for virtual learning environments for those with ASD, however. Burke et
al. (2010) suggest that computer-mediated communication may be useful for individuals with
ASD for both social reasons and practical reasons. Virtual learning environments have recently
begun to be explored as a possible mechanism of social skills instruction for those with ASD.
Even though there are many potential benefits to virtual learning environments, one valid
argument against this is that many virtual learning environments provide a nonsocial
environment for learning that may inhibit generalization of skills to more social environments
(Bishop, 2003). In order to make training more relevant to real-life interactions, Bishop (2003)
suggests that virtual learning environments should be used in collaboration with others.

41

“Telehealth”, “telepsychiatry”, and “virtual therapy” are all terms used to describe the
emerging phenomenon of individuals seeking mental health services on the Internet. There are
multiple pros and cons to this form of service delivery, but one promising aspect is that it may
provide services to those who are unable or unwilling to seek services elsewhere, as the
anonymity of online therapy reduces some of the stigma associated with accessing mental health
services (Stross, 2011). Perceptions of support and specific needs of individuals should be
addressed in all interventions, as perceptions of support are correlated with overall quality of life
in adults with ASD (Renty & Roeyers, as cited in Harmse & Pottas, 2010). Tynes (2007)
suggests that online social networks and virtual communities can be an important source of
psychosocial support for teens. These online communities provide opportunities for adolescents
to seek help and support from others who are like them and ask questions that they may not
otherwise feel comfortable asking someone. Communicating via the Internet can be a unique
form of relationship formation and maintenance that may lead to greater feelings of social
support for individuals. This has not yet been empirically supported for adolescents with ASD,
however, and therefore is worth investigating.
Proposed structure for online implementation of video modeling. There is a lack of
research on best practices for online implementation of video modeling. Drawing on best
practice guidelines for social skills interventions and research on computer-mediated
interventions, however, this researcher has identified specific intervention components that seem
critical for online implementation of video modeling. First and foremost, without an intervention
facilitator it becomes most important that the intervention procedures are specific and easily
understood by the intended audience. The online format must be appropriate for the age and
ability level of the intended recipients and must be easy to navigate. Structure and predictability

42

are important, as well as a clear sequence of lessons. Additionally, since a facilitator is not
present to provide praise or reinforcement for attending to videos, it is important that the online
program be visually appealing, yet not over stimulating. Each of these goals can be
accomplished by selecting an appropriate online platform.
Another key factor is to choose appropriate intervention content. Relevant goals,
multiple learning opportunities, and practice opportunities are important. This can be
accomplished by matching the target skills displayed in the videos to participants’ needs, having
videos that participants can identify with, and providing suggestions for practice opportunities
that may occur in participants’ daily lives. Given the lack of an in-person facilitator, the
immediate practice opportunities that take place in most video modeling interventions are not
possible within an asynchronous online intervention. One way to implement this important
feature of video modeling interventions may be to provide a quiz with video models to check for
understanding after the skill is taught. Quiz feedback may serve a similar purpose and provides a
method for monitoring one aspect of learner performance.
With online implementation of video modeling, another feature that is noticeable absent
is other peers. While some video modeling interventions are implemented independently, others
occur in groups and provide socialization opportunities in addition to instruction. To target this
important feature of face-to-face intervention, an online intervention should provide
opportunities for social interaction between participants. This may help to promote relationship
building, support seeking, or intervention debriefing. Methods for increasing opportunities for
social interaction include discussion forums, instant message, or group chats.
The specific intervention components chosen for the current study, as well as the
rationale for each, are described in the Methods section of this dissertation.

43

Current Study Research Questions and Hypotheses
This literature review has described social skills problems and psychological adjustment
problems for those with ASD, success of previous social skills interventions, the factors that
affect intervention success, and key factors for the development and implementation of an online
intervention. Many social skills intervention strategies have proven to be effective, but high
functioning adolescents with ASD may not always access those interventions and may still be a
challenging group to serve. In addition, adolescents with ASD face potential mental health
challenges, which may or may not be clearly related to their social skills difficulties. In
considering the intervention effectiveness literature and the best practice guidelines for
intervention implementation, it was appropriate to investigate the effectiveness of videomodeling instruction delivered through the Internet to improve social communication skills. It
was also appropriate however, to consider multiple forms of measurement in order to gain a
complete picture of skill development over time. Finally, it was important to further investigate
the relationships between social skills and skill development, and mental health outcomes. To
address these goals, and with the research literature in mind, the following research questions
and hypotheses were written.
This dissertation addresses the following research questions.
Research Question 1. Does an online social skills intervention with video modeling
improve social skills and social support?
It was hypothesized that the intervention group would show greater improvement in
social skills and perceptions of social support over the course of intervention in comparison to
the wait-list control group.

44

Research Question 2. How do parents and adolescents in the intervention group
compare in their views of social skill improvement?
It was hypothesized that parents and adolescents would have different perspectives of the
adolescents’ social skills improvement.
Research Question 3. How do attitudes about the intervention relate to intervention
outcomes?
It was hypothesized that attitudes about the interventions would predict intervention
outcomes (i.e. changes in social and emotional functioning).
Research Question 4A. Are social skills and perceptions of social support predictive of
academic achievement or mental health problems?
It was hypothesized that parent and adolescent reports of social skills and perceptions of
social support would be predictive of mental health problems and academic achievement.
Research Question 4B. Are these relationships different depending on adolescent- or
parent-report?
It was expected that the strength of these relationships would vary depending on reporter.
Research Question 5. Are adolescents’ self-reported changes in social skills related to
changes in mental health problems?
It was hypothesized that changes in social functioning would predict changes in mental
health problems.

45

CHAPTER 3
Research Methods
The online intervention used in the current study builds on a pre-dissertation study by the
author. In this section, the procedures in developing the intervention are first detailed, followed
by the current study procedures.
Pre-Dissertation Study
A pre-dissertation study was conducted to inform the development of the intervention for
the current study. From May 2011 through August 2011, adolescent members of
WrongPlanet.net, an online community for individuals with ASD, were invited to participate in a
research study online. The purpose of this study was to invite adolescent forum members to
share their perceptions of how easy or difficult particular social skills are for them and how
important they deem those skills to be. In addition, adolescents were asked about their preferred
mode of communication with others and their use of online social media. The goal was to gain a
greater understanding of adolescents’ interpersonal skills and social experiences in order to guide
the development of meaningful social skills interventions for higher functioning adolescents with
ASD. The reflections of adolescents with ASD and their identification of relevant challenges
informed the selection of intervention content for the current dissertation study.
Methods. A study announcement was posted in the discussion forum of an online
community for individuals with ASD. The post explained that researchers were interested in
gaining the perspective of adolescents with ASD to better understand their experience. The
researchers directed interested participants to an online survey that included the informed
consent document and a 15-minute survey. The university institutional review board approved

46

the waiver of parental consent to participate in the survey, but adolescents were encouraged to
seek permission from a parent before completing the survey.
The online survey included a demographic questionnaire, communication preferences and
social media questions, and a researcher-developed social skills self-assessment. In addition,
two open-ended questions invited adolescents to identify any other social skills that were
particularly easy or difficult, or any other skills that they felt were important for adolescents to
have.
Results. Preliminary analysis of an initial sample of 49 participants indicated that 24
participants reported their gender as female, 23 participants reported their gender as male, and
two participants did not report their gender. Participants ranged from 13 to 24 years of age (M =
16.76) and all reported currently attending high school or college. Of the 49 respondents, 47
indicated they were high functioning (i.e. they were best described as having high functioning
autism, Asperger’s, or pervasive developmental disorder-not otherwise specified), one indicated
she had a nonverbal learning disability, one indicated she had high-functioning childhood
disintegrative disorder, and none indicated a diagnosis of low functioning autism, childhood
disintegrative disorder, or Rett’s disorder. Thirty five participants indicated that they took all
General Education classes, five indicated they took mostly General Education classes with a few
Special Education classes, five indicated they were homeschooled, one indicated that he took all
Special Education classes, and three described other schedule arrangements (e.g. gifted classes,
testing accommodations). Results indicated that 76% of respondents preferred to communicate
with people their age online and 73% use online social media, such as Facebook.
The social skills identified on the survey were ranked according to mean difficulty ratings
and mean importance ratings to identify where the greatest overlap exists. Identifying the skills

47

that adolescents find both extremely difficult and important is a vital step in enabling educators
and other practitioners to provide targeted, meaningful, and relevant intervention for adolescents
with ASD. Examination of the difficulty ratings and importance ratings, suggested that there are
some skills that adolescents find extremely difficult, but not at all important. For example, on a
scale of 1 (not at all difficult) to 5 (extremely difficult), asking someone to a school dance
received an average rating of 4.58. However, on a scale of 1 (not at all important) to 5 (extremely
important), adolescents gave this particular skill an average rating of 1.86. This indicates that
adolescents recognized that they are not skilled in engaging in this typical teenage behavior, and
it is not very important to them that they are able to do so. There were also skills that
adolescents found extremely difficult and extremely important, however, such as talking to
someone their age about hobbies, understanding sarcasm, and understanding tone of voice, for
example. Overall, this sample of adolescents with ASD highlighted the need for support with
various skills, most of which require social communication strategies. These reflections of
adolescents with ASD and their description of experiences and challenges influenced the
selection of intervention content for this study.
Social Skills Intervention Development
Video model development. Based on this pre-dissertation study, work began to design
the intervention to be used in the dissertation research. Specific social behaviors in the area of
social communication were selected from the results of the pre-dissertation study and from
referencing Social Skills for Teenagers with Developmental and Autism Spectrum Disorders: The
Peers Treatment Manual (Laugeson & Frankel, 2010). The PEERS program was developed at
UCLA’s Semel Institute for Neuroscience and Human Behavior and the curriculum has received
praise from many leaders in the field of ASD research. Additionally, recent literature has

48

provided evidence for the efficacy of the PEERS program (see e.g. Laugeson et al., 2012), and
therefore the intervention manual was thought to be a good resource. This curriculum was
developed for use with teenagers who have ASD and provides materials that target
developmentally appropriate social communication skills. Although the manual was not used to
implement the PEERS program as described, it breaks down complex social skills into specific
behaviors (e.g., entering a conversation) and was used in conjunction with the results of the predissertation study to guide the selection of target behaviors for video modeling (see Table 2
below).
Table 2
Weekly Lesson Topics Adapted From Laugeson & Frankel (2010)
Week 1: Entering a Conversation
Week 2: Exiting a Conversation
Week 3: Two-Way Conversations
Week 4: Beginning Get-Togethers
Week 5: Ending Get-Togethers
Week 6: Handling Disagreements
Week 7: Teasing and Embarrassing Feedback
Week 8: Asking for Help
Video models were developed for each of the target behaviors using the guidelines set
forth by Sansosti, Powell-Smith, and Cowan (2010). Additionally, the researcher sought
additional instruction from How to Use Video Modeling and Video Prompting for the
development of the videos (Sigafoos, O’Reilly, & de la Cruz, 2007). Overall, multiple published
curricula, instructions, and guidelines were consulted in the development of this intervention in
order to draw upon the methods and techniques that have proven successful.
The material from the PEERS curriculum was most closely used as a reference for the
development of video model scripts and copyright permissions were obtained from the authors
and from the Copyright Clearance Center. Scripts were developed for both a male and female to

49

portray each target skill so that participants could view gender specific videos. Additionally,
three separate videos with different scripts were filmed for each skill and each gender, so that
there would be an instructional clip, an example, and non-example, for each of the eight target
skills. The total 48 scripts were sent to a high school drama club that had agreed to film the
videos for the researcher. The drama club was able to successfully complete 44 of the 48 video
clips and the remaining four video clips were filmed by four adolescent acquaintances of the
researcher.
Video models were pre-recorded and streamed online for participants to view. The video
models were short video clips (i.e. 30-180 seconds) that featured adolescents engaging in each of
the target behaviors. A diverse sample of adolescents acted as the models for the videos and
were either provided with community service hours or a small incentive for their assistance.
Website development. The Knoodle online learning management system
(http://www.knoodle.com/) was used to implement and disseminate the intervention to the
intervention group. The Knoodle online learning management system was one of many
commercially available systems to facilitate the development of online learning courses. This
particular system was chosen because of the ease of content development, the opportunity to
integrate video and PowerPoint, the opportunity to create groups within the learning
environment, the social interactivity features, and the testing and reporting analytics that are built
into the system. Additionally, the cost of this particular system was comparable or less than the
cost of other commercially available systems.
Four sample lessons were initially developed on the Knoodle site. Each weekly lesson
consisted of the following components: 1) a rationale for skill development, 2) a first view of
video model clip, 3) prompts for specific interactions to observe within the video, followed by a

50

second view of the video model clip, 4) a review of what was seen in the video, 5) a brief
assessment to check for understanding (i.e., two video clips followed by a quiz asking for the
example of the correct skill), 6) generalization prompts, 7) opportunities for participants to rate
the lesson and provide feedback, and 8) opportunities for participant interaction through
discussion forums. The quiz was presented through the Knoodle quiz function, but all other
components were presented in the form of PowerPoint slides without any audio. Each of the
weekly lessons was designed to be completed in approximately 15-30 minutes, with additional
time required to practice the new skills as suggested in the generalization prompts.
Intervention revisions. After developing the four weekly lessons, a small feedback
session was arranged with adolescents with ASD and their parents to solicit feedback on each of
the proposed elements before intervention implementation occurred. A local parent group was
solicited for feedback and five parents and three of their adolescent children agreed to meet with
the researcher. During a one-hour meeting, the parents and their adolescents viewed the four
lessons and offered suggestions and comments. The researcher took notes and at the end of the
meeting, reviewed the big ideas to ensure that everyone’s voice had been heard. Though
unstructured, this meeting provided valuable feedback as parents suggested the following
primary changes: 1) present one integrated video clip with the quiz embedded, 2) include
voiceover with the text, and 3) add an introductory video to the website that explains how to use
the site features.
After making the changes suggested by parents and adolescents, two members of the
dissertation committee, one of which has video modeling expertise and another who has online
teaching expertise, were consulted for feedback on the draft version of the intervention website.
Additional suggestions for improvement included altering the quality of delivery by completing

51

voiceover recordings with more energy and a slower pace, increasing the font size on the
quizzes, providing differential feedback based on quiz response, and decreasing the video file
size (without losing quality) so that the video would load more quickly. Unfortunately, changing
the font size and providing differential feedback were not possible within the Knoodle quiz
function, but all other changes were made. Also at this time, the remaining four lessons were
completed with the above feedback in mind. The final elements of the online intervention are
displayed in Table 3 below.
Table 3
Elements of Online Intervention
Intervention Element
Introduction to Lesson

Description
Provides a rationale for why the skill is useful or necessary.

Video Clip - First View

Introduces the participant to the video and allows him/her to
watch without interruption.

Video Prompts

Provides questions that prompt viewers to watch for particular
interactions.

Video Clip - Second View

Replays the video to allow participant to watch for interactions
prompted in the previous step.

Review

Provides answers to the questions introduced above.

Brief Skill Assessment

Shows two videos, one example and one non-example, and
instructs participants to choose the correct video. Provides
feedback to participants regarding their learning outcomes.

Suggestions for Practice

Explicitly communicates how to use the learned skills in reallife settings. Also encourages participants to use the
discussion forum to talk about successes and challenges with
practicing skills.

Rating of the Lesson/Feedback

Provides method of assessing participant satisfaction and
hopefully communicates to participants that their feedback is
valued.

Discussion with Peers

Provides opportunity for participants to discuss their successes
and challenges with the target skills or other topics with others.

52

Rationale for intervention components. It is important to provide students with ASD a
rationale for why a particular social skill is useful or necessary because they often have little
social understanding (Myles, 2005). Sansosti et al. (2010) suggest that, “rationales for teaching
social skills should include why the information is important or useful and how the student can
use the information taught within the lesson(s)” (p. 136). Additionally, they suggest that,
“teaching efforts need to provide students with information regarding what skills are going to be
taught, what activities will be used to teach the skills, and the length of time of instruction within
each group session.” Once students with ASD understand why they are participating in social
skills instruction and once they know what to expect, then instruction can begin (Sansosti et al.,
2010).
Suggestions for practice opportunities accompanied each target skill as a means of
encouraging adolescents to use the knowledge they gained. It has been suggested that one of the
greatest drawbacks of virtual interventions is that generalization to real-life settings is less likely
to occur (Bishop, 2003). Therefore, suggestions for role-playing and real-life practice
opportunities were included with each lesson and participants were requested to report whether
or not they followed through with each suggestion in the discussion forums. Additionally, the
discussion forums provided an opportunity for participants to discuss their experiences in
practicing the learned skills.
Participants were invited to provide feedback on each lesson. The user-rating feature of
the online learning management system allowed participants to rate each lesson with one to five
stars and also provide qualitative feedback. This not only provided a method of assessing
participant satisfaction, but also hopefully communicated to participants that their feedback was
valued.

53

Brief quizzes were used to provide feedback to participants so they could track their
progress throughout the intervention. Since there was no instructor present during intervention,
the brief quizzes provided some feedback to participants regarding their learning outcomes.
When appropriate/necessary, there was an opportunity to contact the researcher. Participants
were able to view each lesson multiple times throughout the 8-week intervention and were
allowed to retake quizzes in order to maximize opportunities for success.
Discussion forums were provided as an opportunity for peers to interact with others who
possibly share similar interests, strengths, difficulties, and concerns. Discussion forums provide
a form of asynchronous communication that allows multiple people in geographically distant
locations to interact with one another. It has been found that those with ASD do use technology
to interact with peers, but prefer nonverbal methods of doing so (i.e., cell phones are used for
text messaging more frequently than they are for phone calls; Durkin, Whitehouse, Jaquet,
Ziatas, & Walker, 2010). The discussion forums were monitored, but not moderated, as research
shows that adolescents are likely to engage in more social conversation when the discussion is
not moderated (Oren, Mioduser, & Nachmias, 2002). There was also a mechanism for
participants to report inappropriate behavior to the researcher.
Final considerations. Oren and colleagues (2002) have suggested four guidelines for
establishing virtual learning groups that support the development of a positive social climate:
1) Teachers should moderate the group’s work in a way that enables students to interact.
They should act mainly as facilitators of processes and they minimize their interventions
so as to allow students to gain knowledge from each other and manage discussions
independently. They should therefore refrain from dominating the discussion and from
interacting mainly with individual students (the one -to-many teacher-centric template

54

typical of the frontal classroom), encouraging instead dense student-to-student
interactions.
2) They should also encourage participants to act friendly with each other and create a
relaxed and calm atmosphere.
3) Online course moderators should be attentive to participants’ social needs, and offer a
legitimate platform for messages (or parts in messages) that have social rather than solely
content significance.
4) It is crucial to enhance the social atmosphere by using supportive feedback, discussing
with the group ways to facilitate the creation of social interactions, emphasizing the
importance of peer feedback, and by encouraging students to relate to each other during
the learning activities and beyond (p. 7).
Finally, the following guidelines are suggested by Murray (as cited in Walsh, 2009) for
developing an online environment that can cater to a learner with ASD: reduced stimuli, visual
cues, repetition, clear cut rules, lack of affect, structure, predictability, controllability, interest,
and unhurried pace. Overall, the online learning management system, Knoodle, was selected
based on the above-described guidelines and the above-stated rationales for particular
intervention elements.
Recruitment
Following the completion of the intervention website and approval from the Michigan
State University Institutional Review Board, participants were recruited through the Interactive
Autism Network (IAN) Community (see http://www.iancommunity.org/). This is an online
community, developed at the Kennedy Krieger Institute, designed to link researchers and the
ASD community. The researcher submitted an application to IAN along with the IRB-approved

55

recruitment letter and IAN provided the researcher with the cost for using the recruitment
service. The recruitment fee was paid and IAN then forwarded the recruitment letter to all
members of the IAN Community who met the study’s inclusion criteria (described below). In
total, the advertisement was sent to 1488 parents of 1499 children with ASD, with only 11 emails
that were not deliverable. The advertisement was only sent to families of children between 13
and 18 years of age who did not have a Mental Retardation diagnosis and were spending half or
more of their day in regular education classrooms. Individuals in the IAN database have had their
diagnosis of ASD verified by a review of medical records (Daniels et al., 2011) and clinically
validated (Lee et al., 2010).
Parents and adolescents who were interested in participating contacted the researcher to
be enrolled through an online enrollment form. The online enrollment form asked for parent and
adolescent names, adolescent age and grade, and family contact info to send study materials.
Additionally, there were items to assess whether or not the participant met inclusion criteria so
that only those who qualified were sent further study info. To assess adolescents reading level
(one of the inclusion criteria described below), a brief fifth grade passage was presented as an
example, and parents were asked to indicate whether their child reads above or below a fifth
grade level.
Participants
A power analysis, using G*Power statistical software, was used to establish a minimum
participant sample of goal of 128 total participants. Initially, a recruitment goal of 256 was set,
which would have allowed for up to 50% attrition while still maintaining adequate statistical
power for planned analyses. A total of 89 participants had been recruited and sent pre-test
packets with consent forms to begin the study, but only 50 of those packets were returned when

56

the Knoodle company unexpectedly shut down and announced their website would be closing.
At this time, recruitment was stopped, as any additional participants would not have adequate
time to complete the intervention. Therefore, participants for this study were 50 adolescents with
ASD in grades 9-12, and one parent of each participating adolescent child. All participants were
included in analyses of pre-test data, but participants in the intervention group who did not
complete any of the online lessons were excluded from any analyses that included post-test data.
Both male and female students were included in this study. Though the incidence of
ASD is higher for males, it was believed that female adolescents would benefit from intervention
as well. In the pre-dissertation study described above, there were similar numbers of male and
female respondents suggesting that many females with ASD may be interested in or desire
support in developing social skills. A total of four adolescent females enrolled in the study, but
all were randomly assigned to the wait-list control group.
Students in grades 9-12 were included in this study. Students who were retained and
students who were promoted were not be excluded from the study, but data were examined
during statistical analysis to check for outlying tendencies. Students over 18 years of age were
excluded from the study due to the age limitations of selected measures. Finally, only those who
had regular Internet access and could use a computer independently were included in this study
in order to provide maximum opportunities for participation.
The participant sample was primarily defined according to school placement for the
purposes of this study. Only those who spend more than 50% of their time in general education
were included in the study. Students who are homeschooled were not included in this study,
unless they participated in homeschool groups.

57

The rationale for including only those who spend 50% of their time or more in general
education classes is three-fold. First, as the diagnosis of ASD changes, the diagnostic labels are
less informative for communicating information about one’s particular skill deficits. Students
with diagnostic classifications of Autism, Asperger's, and other PDDs present with various
developmental, language, and cognitive delays, and school placement is a more useful way of
selecting a group of participants that have some shared characteristics in regard to language and
cognitive development.
Second, adolescents with ASD who spend the majority of their time in general education
classrooms are participating in a different social context than those who spend the majority of
their time in special education classrooms. They interact with typically developing peers more
often and are therefore more frequently confronted with misunderstanding typical adolescent
interactions.
Finally, the third reason to use class placement for selecting participants was that those in
a majority of general education classes are less likely to be receiving intensive social supports or
intervention within the schools. As students progress through school grades, there is an
increased focus on academics and less emphasis on social-emotional learning. Despite the fact
that many high functioning adolescents with ASD can perform well and succeed in general
education classes, many still struggle socially and face mental health problems. Although it is
important to keep those students enrolled in general education or advanced placement classes to
increase chances for success beyond graduation, it is also important to provide them with social
and emotional supports. Therefore, this group of participants is an important target group for
online intervention that can be utilized outside of the school setting.

58

An additional, and final, inclusion criterion was that participants must read at a fifth
grade reading level or above. This criterion was established to ensure that any students with
more severe autism symptoms who spend more than 50 percent of their time in general education
due to inclusion policies were excluded from the current study. This also ensured that any
students who had significant reading disabilities, but perhaps still spent significant time in
general education classes, were excluded from the study. It was important to ensure that study
participants had a fifth grade reading level or above so that they were able to successfully read
and interpret lesson instructions, quizzes, and peer discussions. Tables 4-7 display demographic,
diagnostic, school, and intervention information for all included participants.

59

Table 4
Demographic Characteristics of Adolescent Participants
Variable
Gender
Male
Female

Intervention
n
%

Control

Total

n

%

n

%

24
0

100.0
0.0

22
4

84.6
15.4

46
4

92.0
8.0

3
10
3
5
3

12.5
41.7
12.5
20.8
12.5

4
4
10
5
3

15.4
15.4
38.5
19.2
11.5

7
14
13
10
6

14.0
28.0
26.0
20.0
12.0

3
0
18
3

12.5
0.0
75.0
12.5

2
1
21
2

7.7
3.8
80.8
7.7

5
1
39
5

10.0
2.0
78.0
10.0

Age
14
15
16
17
18
Race
Hispanic
Asian
White
Multiracial
Table 5
School Demographic Information for Adolescent Participants
Variable

Intervention
n
%

n

%

n

%

Grade
9
10
11
12

12
8
2
2

50.0
33.3
8.3
8.3

11
9
4
2

42.3
34.6
15.4
7.7

23
17
6
4

46.0
34.0
12.0
8.0

Retained
Yes
No

7
17

29.2
70.8

6
20

23.1
76.9

13
37

26.0
74.0

Promoted
Yes
No

1
23

4.2
95.8

0
26

0.0
100.0

1
49

2.0
98.0

60

Control

Total

Table 6
Diagnostic Information of Adolescent Participants
Variable

Intervention
n
%

Control

Total

n

%

n

%

Age of Diagnosis
2
3
4
5
6
7
8
9
10
11
12
Missing Data

4
3
1
0
2
3
3
2
2
2
1
1

16.7
12.5
4.2
0.0
8.3
12.5
12.5
8.3
8.3
8.3
4.2
4.2

1
6
1
3
1
6
3
2
1
1
1
0

3.85
23.1
3.85
11.5
3.85
23.1
11.5
7.7
3.85
3.85
3.85
0.0

5
9
2
3
3
9
6
4
3
3
2
1

10.0
18.0
4.0
6.0
6.0
18.0
12.0
8.0
6.0
6.0
4.0
2.0

Diagnosis
Autism
Asperger's
PDD-NOS

4
17
3

16.7
70.8
12.5

3
15
8

11.5
57.7
30.8

7
32
11

14.0
64.0
22.0

Additional Diagnosis
Yes
No

21
3

87.5
12.5

21
5

80.8
19.2

42
8

84.0
16.0

Table 7
Intervention Information for Adolescent Participants
Variable

Intervention
n
%

n

%

n

%

Receiving Other Current
Intervention(s)
Yes
No
Missing Data

7
15
2

29.2
62.5
8.3

9
17
0

34.6
65.4
0.0

16
32
2

32.0
64.0
4.0

Received Past
Intervention(s)
Yes
No
Missing Data

14
8
2

58.3
33.3
8.3

17
9
0

65.4
34.6
0.0

31
17
2

62.0
34.0
4.0

61

Control

Total

Measures and Instruments
Multiple measures were selected to gather information about social, emotional, and
academic variables from multiple perspectives at multiple time points. Each measure is
described below and Cronbach’s alpha for each measure is presented in Table 9.
Social skills. Social skills were measured in three different ways: the Social Skills
Improvement System (SSIS; Gresham & Elliot, 2008), the Autism Social Skills Profile (ASSP;
Bellini, 2006), and a researcher-developed Likert scale, referred to as Difficulty Likert from here
forward.
Social Skills Improvement System. The SSIS measures social skills and problem
behaviors from both student and parent perspectives. The rating scale can be administered to
students aged 3-18 and their parents, and can be completed in 10-25 minutes. The SSIS-Student
is written at a second grade reading level and the SSIS-Parent is written at a fifth grade reading
level. The Social Skills domain includes communication, cooperation, assertion, responsibility,
empathy, engagement, and self-control subscales. The competing Problem Behaviors domain
includes externalizing, bullying, hyperactivity/inattention, internalizing subscales. Additionally,
the parent form includes an autism spectrum subscale. This broad assessment of social skills was
administered at Time 1 (i.e. before the experimental group received the intervention) and Time 2
(i.e. after the experimental group received the intervention). Internal consistency reliability is
above .90 for both scales on both the parent and student forms and test-retest reliability is above
.70 for both parent and student forms. Intercorrelations among scales and subscales support
content validity, whereas correlations with other reportedly valid measures (e.g., the Social Skills
Rating System, the Behavior Assessment System for Children, and the Vineland Adaptive
Behavior Scales) support construct validity.

62

The SSIS subscale raw scores are categorized into Behavior Levels rather than converted
to standard scores. For this study, the Problem Behavior subscales were used and therefore the
Behavior Levels in Table 8 are relevant for interpreting the data.
Table 8
SSIS Problem Behavior Subscale Raw Scores Corresponding to Behavior Levels
Parent Report
Subscale

Adolescent Report

Below
Average

Average

Above
Average

Below
Average

Average

Above
Average

Externalizing

---

0-10

11-36

0

1-14

15-36

Bullying

---

0-3

4-15

---

0-5

6-15

Hyperactivity/
Inattention

---

0-7

8-21

0

1-11

12-21

Internalizing

---

0-9

10-30

0

1-13

14-30

Autism Spectrum

0-1

2-13

14-45

---

---

---

Note. From Gresham & Elliot, 2008, p. 201.
Autism Social Skills Profile. The ASSP was designed as a parent- or teacher-report
instrument for use with children and adolescents with ASD. In particular, this scale was
developed for use as an intervention-planning tool and as a tool for monitoring intervention
progress when working with youth with ASD. The 49 items on the ASSP represent skills in
initiation, reciprocity, perspective taking, and non-verbal communication. This measure of
social skills was administered to parents at Time 1 and Time 2, in addition to the SSIS, as it was
believed to potentially be a better measure of change over the course of an 8-week intervention.
An analysis of the psychometric properties of the ASSP revealed high internal consistency (.92)
and high test-retest reliability (.90) (Bellini & Hopf, 2007). Additionally, principal component
analysis revealed a three-factor model that included Social Reciprocity, Social

63

Participation/Avoidance, and Detrimental Social Behaviors subscales (Bellini & Hopf, 2007).
Difficulty Likert scales. These scales were developed by the researcher to assess
participants’ perceived difficulty with each of the target intervention skills. Adolescent
participants were instructed, “Please rate how difficult you think the following skills are.” A
scale from 1 (not at all difficult) to 7 (very difficult) was presented and each of the target skills
for intervention was listed. Parents were presented with the same Likert scale and same items,
but prompted, “Please rate how difficult you think the following skills are for your child”. The
purpose of these scales was to assess participants’ changes in perceived difficulty with the exact
skills that were being taught in the intervention. Although the validity of this approach is limited
methodologically, it aligns well with this study’s theoretical foundation and provided unique
information about participant perspectives, as well as more specific information about the
particular target skills.
Table 9
Current Study Scale Reliability
Study Measure
SSIS Child Social Skills Total Score
SSIS Child Problem Behaviors Total Score
SSIS Parent Social Skills Total Score
SSIS Parent Problem Behaviors Total Score
SSIS Parent Autism Spectrum (Subscale)
RCMAS-2 Total Score
RADS-2 Total Score
ASSP Total Score
Child Difficulty Likert Ratings
Parent Difficulty Likert Ratings
Child MSPSS Total Score
Parent MSPSS Total Score
Perceived Competence for Learning Scale
IMI – Interest/Enjoyment
Attitude Toward Using Technology

64

Cronbach’s Alpha
.92
.80
.94
.78
.49
.89
.88
.91
.84
.81
.90
.86
.81
.95
.85

Perceived social support. Parents’ and adolescents’ perceptions of adolescents’ social
support were measured using the Multidimensional Scale of Perceived Social Support (MSPSS;
Zimet, Dahlem, Zimet, & Farley, 1988). The scale consists of 12 items that are answered on a 7point rating scale. The 12 items can be divided into three factor groups that address the source of
social support: family, friends, and significant other. The scale is brief, is written at a fourth
grade level, and improves upon other scales that only assess support from family and friends
(e.g., Procidano & Heller, 1983) making it a useful tool for conducting research with various
groups. Though initially developed using a sample of university students, the MSPSS has been
widely tested and used across diverse samples, including adolescents (see Canty-Mitchell &
Zimet, 2000; Cheng & Chan, 2004). The authors of the scale report internal consistency of .88
(.91 for significant other, .87 for family, and .85 for friends) and test-retest reliability of .85 (.72
for significant other, .85 for family, and .75 for friends). Internal consistency of the MSPSS was
strong when examined in a prior unpublished study (i.e., this author’s thesis) of 12 adolescents
with ASD (.88 total, .81 for significant other, .95 for family, and .97 for friends). Construct
validity of the MSPSS was assessed during scale development based on the hypothesis that
perceived social support would be negatively related to reported depression and anxiety
symptoms. Zimet et al., 1988 found that perceived social support was inversely related to
reported depression on the Hopkins Symptoms Checklist (Derogatis, Lipman, Rickels,
Uhlenhuth, & Covi, 1974) at the p = < .01 level for all three constructs and for the scale as a
whole. Additionally, they found that perceived social support from family was inversely related
to reported anxiety on the Hopkins Symptom Checklist (p = < .01).
This self-report scale was slightly adapted for use as a parent-report survey as well. Each
item of the questionnaire was altered to allow parents to provide their perceptions of the

65

adolescent’s social support. This adaptation was made in a prior unpublished study (i.e., this
author’s thesis), and internal consistency remained high (.86 total, .87 for significant other, .92
for family, and .97 for friends). In the current study, reliability estimates of the parent measure
were obtained before statistical analysis was completed to ensure that the scale maintained
adequate internal-consistency reliability despite the adaptation (see Results).
Depression. Participants were asked to complete the Reynolds Adolescent Depression
Scale, Second Edition (RADS-2; Reynolds, 2002) at Time 1 and Time 2. The RADS-2 is a brief
screener for adolescent depression that includes 30 items written at a third grade level. It can be
administered to adolescents aged 11 to 20 and completed in five to 10 minutes. The items cover
dysphoric mood, anhedonia/negative affect, negative self-evaluation, and somatic complaints.
The RADS-2 yields a total score as well as clinical cutoff scores that allow evaluation of
symptom severity into normal, mild, moderate, and severe categories. Internal consistency
reliability for Depression Total is .94, whereas test-retest reliability is .85. Content and construct
validity are supported by numerous empirical studies detailed in the professional manual.
Anxiety. Anxiety was measured using the Revised Children’s Manifest Anxiety Scale,
Second Edition (RCMAS-2; Reynolds & Richmond, 2008) at Time 1 and Time 2. The RCMAS2 is a self-report inventory that allows for the screening of anxiety in children from six to 19
years old. It is written in a simple yes-no response format and can be completed in
approximately 10 minutes. The RCMAS-2 consists of 49 items covering scales of physiological
anxiety, worry, social anxiety, and defensiveness, as well as an inconsistent responding index.
Internal consistency reliability for Total Anxiety is .92, whereas test-retest reliability is .76.
Interscale correlations were moderate to high suggesting adequate content validity.
Demographic information. Parents were asked to complete a demographic

66

questionnaire (see Appendix B) that includes items for the adolescent participants’ gender, age,
grade, race, relevant educational history, diagnostic information, and relevant intervention
history. Additionally, on this demographic survey, parents were asked to report adolescents’
current grades and GPA as measures of academic achievement. GPA was reported on a 4.0
scale, whereas grades were reported as Mostly As, Mostly Bs, Mostly Cs, Mostly Ds, or Mostly
Fs and then converted to a 5-point ordinal scale. If more than one grade category was selected,
the median score was used for analysis.
Attitudes. To examine the influence of various contextual factors on adolescents’
experience using the online intervention, three different scales were used to assess the attitudes
of participants in the intervention group. Each of these scales measuring attitudes was slightly
adapted for the current study, and therefore reliability of each scale was assessed with the current
study sample prior to conducting any statistical analyses.
Intrinsic Motivation Inventory. The perceived competence and interest/enjoyment
subscales of the Intrinsic Motivation Inventory (IMI; Ryan, 1982) were used to assess
participants perceived competence for learning social skills and their enjoyment of the online
intervention. The IMI is a multidimensional measure that is used to assess participants’
subjective experiences of an activity. It has been used in many research studies across
disciplines, including education, and has been validated in several experiments (see
selfdeterminationtheory.org for a list of selected publications).
Attitude Toward Using Technology Scale. Finally, participants’ attitudes toward the
online course management system (i.e. Knoodle) were assessed with a 4-item Likert scale
(Venkatesh, Morris, Davis, & Davis, 2003). This scale, although not often used in intervention
studies, has strong research support in the communications and information technology

67

literature, as it is an integral part of the well-known Technology Acceptance Model (Venkatesh
& Davis, 2000).
Exposure. The number of weekly lessons completed was used to measure participation
and was recorded by the online learning management system. This also served as a measure of
treatment integrity to ensure that participants in the intervention group were in fact viewing the
intervention lessons.
Participant satisfaction. Additionally, participant ratings of the online intervention
lessons were used to examine participant satisfaction. The user-rating feature of the online
learning management system allowed participants to rate each lesson with one to five stars and
also provide qualitative feedback.
Procedure
Time 1 assessment. All enrolled participants were sent a pre-test packet following
enrollment in the study. Though online pre-test measures would have been ideal, it was not
possible to use copyrighted, standardized measures online. The pre-test packet included a
welcome letter, consent form, assent form, a demographic questionnaire, and all Time 1
measures (i.e., SSIS, ASSP, MSPSS, Difficulty Likert scales, Perceived Competence for
Learning scale, RADS-2, RCMAS-2). Parents and adolescents were instructed to complete and
return all documents within a 1-week period. Upon receiving a complete packet, the researcher
provided the participant with a $10 incentive.
Implementation. As pre-test packets were received, participants were assigned to either
the online video modeling intervention or a wait-list control group using random assignment.
More specifically, a random number generator was used to assign the participants to two

68

conditions. One group was asked to begin the 8-week intervention immediately and the other
was put on a waiting list.
The intervention began with a brief welcome in Week 1 to explain the purpose and
structure of intervention. Additionally, a participation agreement form that detailed the norms
and expectations for communicating in the discussion forums was shared with participants. The
purpose of this agreement was to explicitly communicate the social norms of online
communication and to minimize the risk of bullying among participants. Participants had to
agree to abide by the proposed expectations before beginning the intervention. Following the
introduction, each weekly lesson was focused on social communication skills (see Table 2).
While the intervention was in progress, Knoodle suddenly announced the closing of their
company. The researcher received a two-week notice of the website shutting down and
therefore, the best solution was to have all currently enrolled intervention participants to
complete a 6-week intervention rather than the intended 8-week intervention (i.e. lessons seven
and eight were dropped from the intervention). After an immediate, careful review of
alternatives, a replacement website, Mindflash (http://www.mindflash.com/home) was promptly
selected for the control group. Mindflash is a similar course management system that provides a
user-friendly interface, opportunities for video streaming, PowerPoint presentation, quizzes,
surveys, and reporting analytics. Unlike Knoodle, Mindflash does not provide options for social
interaction between users. However, also unlike Knoodle, Mindflash does provide opportunities
for differentiated feedback within the quiz function. Most importantly, Mindflash allowed for
participant anonymity and a site that could be developed with a similar structure and in a timely
manner.

69

Time 2 assessment. Immediately following the conclusion of intervention, post-test
measures (i.e., SSIS, ASSP, Difficulty Likert scales, RADS-2, RCMAS-2, and attitudes scales)
were mailed to adolescents and their parents in both groups. Similar to pre-test packets, it was
requested that materials were returned within a 1-week period. After post-test packets were
returned, $10 incentives were sent to participants. At that time, participants in the wait-list
control group began the intervention on the Mindflash site. Participants in both groups who
participated fully throughout the duration of the study were entered in a raffle for one of two
pocket video cameras at the conclusion of the study.

70

CHAPTER 4
Results
All raw participant data were entered into IBM SPSS Statistics software on a passwordprotected computer. Identification numbers assigned to each participant were entered into the
database; no identifying information was recorded in the data set. The researcher wrote syntax to
calculate total scores (e.g., standard scores, T-scores, and difference scores) for all relevant
variables in the data set. Intervention metadata and preliminary data checks are discussed first,
followed by the results of each of the research questions. Due to the large number of measures
utilized in this study, and the various statistical procedures used across research questions,
preliminary analyses and descriptive statistics are presented separately for each research
question.
Intervention Metadata
In order to appropriately analyze the study data, it is first important to look at the
metadata, which provide a better understanding of the extent to which participants utilized the
intervention components. In Table 10 below, data regarding lessons completed, posts in the
discussion forum, and lesson ratings are presented.
The data show that on average, participants completed just over two thirds of the
available lessons. Of those lessons completed, participants were able to answer the quiz
questions correctly in approximately 97 percent of attempts made. Lessons were rated an
average of 3.8 out of five stars, and all six lessons were rated similarly.
Unexpectedly, the discussion forum was utilized infrequently and only four of the 24
intervention participants posted at all. Although a rationale was provided in the introduction to
this dissertation for examining online communities, and research literature suggests positive

71

social-emotional effects for virtual communities, it is not appropriate to examine such effects
within this study given the lack of use of the discussion forum component. As such, the research
questions below will not address changes in social support over time. Those changes were
expected to result from the opportunities provided by the discussion forum, and participants did
not pursue those opportunities.
Table 10
Intervention Metadata
Intervention Participants

N = 24

Lessons Completed
(6 Possible Lessons)

M = 4.5

Quiz Scores

Average % Correct: 97.2
Average % Incorrect: 2.8

Lesson Ratings
(5 Stars Possible)
Users Completing Weekly
Ratings

M = 3.8; Range 3.7-4
M = 7.8; Range 5-14

Discussion Posts

N=9

Users Posting in Forum

N=4

Preliminary Data Checks
To ensure accurate data entry and syntax accuracy, a minimum of ten percent of each
measure was also hand-scored and compared to the SPSS dataset to check for inaccuracies. Any
inaccuracies that were found were corrected by comparing the SPSS database with the raw data
to identify errors. Additionally, raw data were visually examined and descriptive statistics were
used to check for errors. No more than two errors were found across all measures.
Prior to examining the results of any analyses, reliability checks were completed for each
of the study measures. For the purposes of this study, reliability above 0.70 was considered

72

acceptable, as suggested by George and Mallery (2003). As seen in Table 8, all measures had
reliabilities at or greater than 0.70 except for the Autism Spectrum subscale of the Parent SSIS
measure.
Research Question 1: Does an online social skills intervention improve social skills?
The primary goal of this study was to examine whether or not an online social skills
intervention with video modeling could improve social skills. Social skills were measured
through parent- and adolescent-report on the SSIS, parent-report on the ASSP, and through a
Likert scale that measured perceptions of social skill difficulty from both parent- and adolescentreport.
Parent perspectives were examined first, and a MANCOVA was run to examine the
differences in social skills between groups at post-test, while controlling for pre-test scores.
Three measures of social skills were used to assess parent perspectives on adolescents’ social
skills: SSIS scores, ASSP scores, and Difficulty Likert scores. Preliminary checks of
assumptions indicated that data was normally distributed, as assessed by Shapiro-Wilks test (p >
.05). Also, there were no univariate or multivariate outliers, as assessed by a visual inspection of
boxplots. There were linear relationships between variables, as assessed by visual inspection of
scatterplots, and Box’s M test indicated there was homogeneity of variance-covariance matrices.
Due to multicollinearity between the ASSP and the SSIS (r = .850, p = < .01), the ASSP measure
was dropped from the analysis. The resulting model included two measures of social skills, the
SSIS and the Difficulty Likert scale. As shown in Figure 1, participants in the intervention group
scored higher on the SSIS (M = 80.39, SD = 13.44) and lower on the Difficulty Likert scale (M =
27.39, SD = 7.45) than the wait list group (M = 77.32, SD = 15.32; M = 32.91, SD = 7.39). The

73

difference between groups on the combined dependent variables was not significant, however (F
2

= 1.181, p = .318, Wilks' Λ = .938, partial η = .062).
Figure 1
Mean Social Skills Scores at Time 2 from Parent Report
SSIS

Difficulty Likert

85

33.5

70

26.5
Intervention

Wait List Control

Intervention

Wait List Control

Note. SSIS Social Skills scores have a mean of 100 and a standard deviation of 15. Scores
between 70 and 85 fall within the Below Average range. The Difficulty Likert scale had a
standard deviation of approximately 7 within the study sample. Possible scores ranged from
seven to 42 with higher scores indicating greater difficulty.
A second MANCOVA was run to examine the differences in social skills between groups
at post-test, while controlling for pre-test scores, from adolescent perspectives. Similar to the
first MANCOVA, two measures of social skills were utilized: SSIS scores and Difficulty Likert
scores. The same assumptions were checked for this analysis and data supported the use of the
MANCOVA procedure. As shown in Figure 2, participants in the intervention group scored
higher on the SSIS (M = 96.19, SD = 17.29) and lower on the Difficulty Likert scale (M = 19.13,
SD = 6.28) than the wait list group (M = 92.68, SD = 17.40; M = 22.18, SD = 9.28). The

74

difference between groups on the combined dependent variables, however, was not significant, F
2

= .953, p = .448, Wilks' Λ = .953, partial η = .047.
Figure 2
Mean Social Skills Scores at Time 2 from Adolescent Report
SSIS

Difficulty Likert

100

23.5

85

16.5
Intervention

Wait List
Control

Intervention

Wait List
Control

Note. SSIS Social Skills scores have a mean of 100 and a standard deviation of 15. Scores
between 70 and 85 fall within the Below Average range, and scores between 85 and 115 fall
within the Average range. The Difficulty Likert scale had a standard deviation of approximately
7 within the study sample. Possible scores ranged from seven to 42 with higher scores indicating
greater difficulty.
Due to the lack of clear differences in social skill changes between the two groups, the
problem behaviors scale of the SSIS was also examined to see if any trends would emerge when
skill deficits were considered instead. Given that problem behaviors often emerge for those with
ASD due to individuals’ lack of social skills, a secondary hypothesis was developed during data
analysis that perhaps problem behaviors had decreased for the intervention group even though
clear social skill improvement had not been attained.
A MANCOVA was run to examine the differences in problem behaviors between groups
at post-test, while controlling for pre-test scores. Five subscale measures were used to assess

75

parent perspectives on adolescents’ problem behaviors: Externalizing, Internalizing, ASD
Symptoms, Bullying, and Hyperactivity/Inattention subscales. Assumptions of normality,
linearity, homogeneity of variance-covariance matrices, and multicollinearity were satisfied.
Additionally, box plots were used to conduct a visual inspection of univariate and multivariate
outliers and none were found. As displayed in Figure 3, participants in the intervention group
scored lower on all five subscales than the wait list group. The difference between groups on the
2

combined dependent variables was significant, F = 3.007, p = .026, Wilks' Λ = .666, partial η =
.334. Follow-up univariate analyses were conducted to examine the treatment effects on each of
the individual subtests. There were statistically significant differences between groups on the
2

Internalizing (F = 8.695, p = .006, partial η = .204), ASD Symptoms (F = 4.824, p = .035,
2

2

partial η = .124), Bullying (F = 4.658, p = .038, partial η = .120), and Hyperactive/Inattention
2

subscales (F = 5.496, p = .025, partial η = .139, but not on the Externalizing subscale (F =
2

2.878, p = .099, partial η = .078), with the intervention group reporting fewer problem
behaviors.

76

Figure 3
Mean Problem Behavior Subscale Scores at Time 2 from Parent Report
Intervention

Wait List Control

25
20
15
10
5
0

Note. SSIS subscale scores are categorized into “Behavior Levels” rather than converted to
standard scores. See Table 8 in the Methods section for descriptive categories.
Once again, a similar analysis was run to examine the differences in problem behaviors
between groups from adolescent perspectives. This time, however, four subscale measures were
used (i.e. Externalizing, Internalizing, Bullying, and Hyperactivity/Inattention subscales) because
the adolescent report measure does not include the ASD Symptoms subscale. The same checks
for assumptions were conducted and the use of the MANCOVA was procedure was supported.
Similar to parents, adolescents in the intervention group rated their problem behaviors lower on
all four subscales compared to the wait list group (see Figure 4). In contrast to the parent report
results, the difference between groups on the combined dependent variables from adolescent
2

perspectives was not significant (F = .576, p = .682, Wilks' Λ = .933; partial η = .067).
77

Figure 4
Mean Problem Behavior Subscale Scores at Time 2 from Adolescent Report
Intervention

Wait List Control

10

5

0

Note. SSIS subscale scores are categorized into “Behavior Levels” rather than converted to
standard scores. See Table 8 in the Methods section for descriptive categories.
Since the SSIS Problem Behaviors Scale has greater reliability than each of its subscales,
and because the MANCOVA analysis had limited power, this researcher decided to also conduct
an ANCOVA using the Problem Behaviors post-test total score as the dependent variable and
controlling for pre-test scores. As with the MANCOVA analyses above, linearity was checked
through visual inspection of a scatterplot, normality was assessed by Shapiro-Wilk’s test (p <
.05), Levene’s Test of Homogeneity of Variance was satisfied (p < .05), and no outliers were
found through visual inspection of boxplots. The results displayed in Figure 5 show that from
parent perspectives, there was a statistically significant difference between intervention and wait2

list groups on post-test measures of problem behaviors (F = 7.936, p = .008, partial η = .173),

78

with the intervention group reporting fewer problem behaviors (M = 102.32, SD = 5.89) than the
wait list group (M = 110.36, SD = 11.36).
Figure 5
Mean Problem Behavior Total at Time 2 from Parent Report
115

100
Intervention

Wait List Control

Note. SSIS Problem Behavior scores have a mean of 100 and a standard deviation of 15. Scores
between 85 and 115 fall within the Average range.
Similarly, a repeated measures ANCOVA examining adolescent perspectives was
conducted using the Problem Behaviors total score. All relevant assumptions were checked and
satisfied. As displayed in Figure 6, the intervention group reported fewer problem behaviors (M
= 92.89, SD = 6.87) than the wait list group (M = 98.64, SD = 9.42). The ANCOVA indicated
2

that this group difference approached significance (F = 3.799, p = .059, partial η = .091).

79

Figure 6
Mean Problem Behavior Total at Time 2 from Adolescent Report
100

85
Intervention

Wait List Control

Note. SSIS Problem Behavior scores have a mean of 100 and a standard deviation of 15. Scores
between 85 and 115 fall within the Average range.
All descriptive and statistical data for the MANCOVA and ANCOVA analyses described above
are displayed in Table 11.

80

Table 11
Treatment Effects
Intervention
M
SD

Wait List Control
n
M
SD

F
P Value
Statistic

Partial
2
η

n
SOCIAL SKILLS
1.181
.318
.062
Parent Report
SSIS Total
18
80.39
13.44
22
77.32
15.32
------Difficulty Likert
18
27.39
7.45
22
32.91
7.39
------SOCIAL SKILLS
.953
.448
.047
Adolescent Report
SSIS Total
16
96.19
17.29
22
92.68
17.40
------Difficulty Likert
16
19.13
6.28
22
22.18
9.28
------PROBLEM BEHAVIOR SUBSCALES
3.007
.026*
.334
Parent Report
Internalizing
19
6.68
2.95
22
11.82
6.37
8.695
.006**
.204
Externalizing
19
9.00
4.62
22
10.59
6.48
2.878
.099
.078
Hyperactivity/Inattention
19
7.74
3.26
22
9.41
3.47
5.496
.025*
.139
Bullying
19
1.42
1.35
22
2.55
2.97
4.658
.038*
.120
Autism Spectrum
19
19.63
5.39
22
21.73
6.20
4.824
.035*
.124
PROBLEM BEHAVIOR SUBSCALES
.576
.682
.067
Adolescent Report
Internalizing
19
4.95
3.57
22
9.00
6.17
------Externalizing
19
6.95
4.29
22
8.59
4.81
------Hyperactivity/Inattention
19
7.42
4.21
22
7.86
3.50
------Bullying
19
1.53
1.93
22
2.00
1.77
------PROBLEM BEHAVIOR TOTAL
Parent Report
19
102.32
5.89
22
110.36
11.36
7.938
.008**
.173
Adolescent Report
19
92.89
6.87
22
98.64
9.42
3.799
.059
.091
** p < .01; * p < .0; Results of MANCOVA and ANCOVA results are in bold and follow-up univariate analyses are in plain text.
81

Research Question 2: How do parents and adolescents in the intervention group compare
in their views of social skill improvement?
A second goal of this study was to examine how outcomes for those in the intervention
group varied by reporter. In particular, differences between adolescent and parent report were
examined on a standardized measure of social skills (i.e. SSIS) and through a Likert rating of
social skill difficulty. Although not an initial goal of the study, differences in problem behavior
change over time were examined by reporter as well, given the significant decrease in problem
behavior for the intervention group.
An initial inspection of mean differences at pre-test was conducted prior to examining
reporter-time interaction effects. Pre-test data sets contained no outliers and the data were
normally distributed, as assessed through visual inspection of boxplots and Normal Q-Q plots. A
paired samples t-test indicated a mean SSIS standard score difference of 10.667 between
adolescent and parent ratings, with adolescents reporting significantly higher social skill scores (t
= 2.992, p = .007, d = 0.65). Similarly, a paired samples t-test indicated a mean Difficulty Likert
score difference of 9.762 between adolescent and parent ratings of social skill difficulty at pretest. This is a significant difference between reporters, with parents reporting greater difficulty (t
= -5.524, p < .001, d = 1.21).
To examine differences in perspectives on social skills change throughout the
intervention, a repeated measures MANOVA procedure was used. Two measures of social skills
were used with both adolescent and parent reporters to assess perspectives on adolescents’ social
skills: SSIS scores and Difficulty Likert scores. Preliminary checks of assumptions indicated that
the data were normally distributed (i.e. Shapiro-Wilk’s test p > .05). A visual inspection of
boxplots indicated that there was one univariate outlier in the child SSIS data (i.e. one data point

82

that was more than 1/5 box-lengths from the edge of the box in the boxplot). Inspection of the
value did not reveal it to be extreme and it was retained in the analysis. All other adolescent and
parent data contained no univariate or multivariate outliers (i.e. Mahalonobis distance p > .001).
Linear relationships between variables were confirmed through a visual inspection of
scatterplots, and Box’s M test revealed there was homogeneity of variance-covariance matrices.
Additionally, multicollinearity was not found upon examination of Pearson correlations between
variables.
When examining social skills change over time, there was no statistically significant
difference between reporters on the combined dependent variables (i.e. SSIS and Difficulty
2

Likert scores), F = 1.805, p = .181, Wilks' Λ = .899; partial η = .101 (see Figures 7 and 8).
Though the interaction between reporter and time was not significant, there was a main effect of
2

reporter (F = 6.836, p = .003, Wilks' Λ = .701; partial η = .299), with parents reporting weaker
social skills (i.e. lower SSIS scores and higher Difficulty Likert scores) at both times.

83

Figure 7
SSIS Scores at Time 1 and Time 2 for Parent and Adolescent Report
Parent

Adolescent

100

85

70
Time 1

Time 2

Note. The SSIS has a mean of 100 and a standard deviation of 15. Scores between 70 and 85
fall within the Below Average range, and scores between 85 and 115 fall within the Average
range.
Figure 8
Difficulty Likert Scores at Time 1 and Time 2 for Parent and Adolescent Report
Parent

Adolescent

28

21

14
Time 1

Time 2

Note. The Difficulty Likert scale had an average standard deviation of 7 across reporters.
Possible scores ranged from seven to 42 with higher scores indicating greater difficulty.

84

To examine reporter differences in perspectives on problem behavior changes throughout
the intervention, a repeated measures MANOVA procedure was also used. The four subscales
present on both the adolescent and parent measures were used for analysis: Externalizing,
Internalizing, Bullying, and Hyperactive/Inattentive. The same checks of assumptions conducted
for the previous MANOVA were repeated and all assumptions were met. Results of the repeated
measures MANOVA indicated that there was a non-significant reporter-time interaction on the
2

combined dependent variables, F = .673, p = .616; Wilks' Λ = .925; partial η = .075. There was
a main effect for time however, with both parent and adolescent reporting fewer problem
2

behaviors at Time 2 (F = 8.56, p < .001, Wilks' Λ = .490; partial η = .510). Follow-up
univariate analyses indicated the main effect of time was significant for all four subscales:
Hyperactivity/Inattention (p = .029), Bullying (p = .032), Externalizing Problems (p = .020), and
Internalizing problems (p < .001). See Figures 9-12 for a visual display of problem behavior
changes over time from both reporters.

85

Figure 9
Hyperactivity/Inattention Scores at Time 1 and Time 2 for Parent and Adolescent Report
Parent

Adolescent

10.5

6.5
Time 1

Time 2

Note. The average standard deviation across reporters was approximately 4.0. See Table 8 for
SSIS Problem Behavior subscale Behavior Levels (i.e. descriptive categories).
Figure 10
Bullying Scores at Time 1 and Time 2 for Parent and Adolescent Report
Parent

Adolescent

2.6

0.9
Time 1

Time 2

Note. The average standard deviation across reporters was approximately 1.7. See Table 8 for
SSIS Problem Behavior subscale Behavior Levels (i.e. descriptive categories).

86

Figure 11
Internalizing Symptoms Scores at Time 1 and Time 2 for Parent and Adolescent Report
Parent

Adolescent

10.1

6.7

3.3
Time 1

Time 2

Note. The average standard deviation across reporters was approximately 3.4. See Table 8 for
SSIS Problem Behavior subscale Behavior Levels (i.e. descriptive categories).
Figure 12
Externalizing Symptoms Scores at Time 1 and Time 2 for Parent and Adolescent Report
Parent

Adolescent

10.8

6
Time 1

Time 2

Note. The average standard deviation across reporters was approximately 4.8. See Table 8 for
SSIS Problem Behavior subscale Behavior Levels (i.e. descriptive categories).

87

Given the lack of a significant interaction effect at the subtest level, a repeated measures
ANOVA was used to also examine reporter differences over time on the problem behaviors total
score. Once again, all relevant assumptions were checked and met. Results of the repeated
measures ANOVA (shown in Figure 13) indicated that there was not a significant difference
between adolescent and parent reporter changes in problem behavior scores from Time 1 to Time
2

2 (F = .000, p = 1.000, Wilks' Λ = 1.000; partial η = .000). There were however, main effects
2

for time (F = 18.622, p < .001, Wilks' Λ = .659; partial η = .341) and for reporter (F = 21.480, p
2

< .001, partial η = .374).
Figure 13
Problem Behavior Total Scores at Time 1 and Time 2 for Parent and Adolescent Report
Parent

Adolescent

115

100

85
Time 1

Time 2

Note. SSIS Problem Behavior total scores have a mean of 100 and a standard deviation of 15.
Research Question 3: How do attitudes about the intervention relate to intervention
outcomes?
To complement research questions one and two, this question investigated possible
88

predictors of intervention effectiveness. More specifically, perceived competence for learning
(as measured by the PCLS), interest and enjoyment (as measured by the IMIInterest/Enjoyment), and attitudes toward using the online intervention (as measured by the
Attitude Toward Using Technology Scale), were hypothesized to be predictors of intervention
effectiveness. Perceived competence for learning was measured before the intervention began,
and interest/enjoyment and attitudes toward the online intervention were measured after the
intervention had concluded. The primary outcome variable of interest was social skills, which
was measured in multiple ways (i.e. SSIS-Child, SSIS-Parent, Child Likert, Parent Likert,
ASSP). For this analysis, gain scores representing the change from Time 1 to Time 2 were used
as outcome variables. Additionally, changes in adolescent reports of anxiety (i.e. RCMAS gain
scores) and depression (i.e. RADS gain scores) were also outcomes of interest, as it was
hypothesized that opportunities for skill development and for relationship building (through the
online forums) may have secondary effects on adolescents’ emotional outcomes as well. To
examine the influence of attitudes on each of these outcome variables, multivariate regression
was used.
Prior to performing each analysis, independence of residuals was assessed by examining
Durbin-Watson statistics, all of which ranged from 1.5-2.3 indicating the assumption of
independence was met. Additionally, for each analysis, the assumptions of linearity,
independence of errors, homoscedasticity, and normality of residuals were met.
First, a multiple regression was performed to predict social skills gain, as rated by
adolescents (N=17) on the SSIS, from perceived competence for learning, interest/enjoyment,
and attitudes toward using technology. These variables did not significantly predict social skills
2

gain as rated by adolescents, F = .108, p = .954, R = .024. None of the three variables added

89

statistically significantly to the prediction. Next, a multiple regression was run to predict social
skills gain, as rated by parents (N=17) on the SSIS, from perceived competence for learning,
interest/enjoyment, and attitudes toward using technology. Attitudes also did not significantly
2

predict social skills gain as rated by parents on the SSIS, F = 1.025, p = .414, R = .191. None of
the three variables added significantly to the prediction. Additionally, attitudes did not
2

significantly predict social skills gain on the ASSP, F = .848, p = .494, R = .175. None of the
three variables added significantly to the prediction.
Next, decreases in difficulty ratings on the Likert scale were examined. A multiple
regression was run to predict decrease in social skill difficulty ratings, as rated by adolescents
(N=16) on the Likert scale, from perceived competence for learning, interest/enjoyment, and
attitudes toward using technology. These variables did not significantly predict difficulty
2

changes reported by adolescents, F = .967, p = .440, R = .195. None of the three variables added
significantly to the prediction. Similarly, attitudes did not significantly predict difficulty changes
2

rated by parents (N=17), F = .348, p = .791, R = .074. None of the three variables added
significantly to the prediction.
Finally, to examine the potential influence of attitudes on mental health changes, anxiety
and depression gain scores were used. A multiple regression was run to predict depression
changes as rated by adolescents (N=17) on the RADS, from perceived competence for learning,
interest/enjoyment, and attitudes toward using technology. These variables did not significantly
2

predict depression changes, F = 2.027, p = .160, R = .319. None of the three variables added
significantly to the prediction. Similarly, these variables did not significantly predict anxiety

90

2

changes as measured by the RCMAS, F = .309, p = .818, R = .067. None of the three variables
added significantly to the prediction.
Research Question 4A: Are social skills and perceptions of social support related to
academic achievement or mental health problems?
An additional goal of this study was to investigate the relationship between social skills
and perceptions of social support, and academic (i.e. grades and GPA) and mental health
outcomes (i.e. anxiety and depression; see Table 12 for correlations between Time 1 measures,
and Table 13 for correlations between Time 2 measures).

91

Table 12
Pearson Correlations Between Pre-Test Measures
1.
1. PCLS

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

1

2. MSPSS-A

.388*

1

3. MSPSS-P

.264

.424**

1

-.260

-.087

.060

1

-.231

-.394**

-.270

.086

1

.365*

.441**

.125

.012

-.321*

1

.177

.034

.195

.095

.278

-.198

1

.030

.213

.498**

.180

-.504**

.379**

-.262

1

-.103

-.167

-.316*

-.236

.404**

-.509**

.311*

-.737**

1

10. ASSP

-.027

.158

.473**

.157

-.511**

.276

-.181

.850**

-.627**

1

11. RADS-2

-.146

-.263

-.220

.090

.415**

-.240

.514**

-.291*

.427**

-.193

1

12. RCMAS-2

-.038

-.067

.213

.366*

.444**

-.017

.656**

-.031

.106

.086

.633**

4. Difficulty
Likert-A
5. Difficulty
Likert-P
6. SSIS-A
Soc Skills
7. SSIS-A
Prob Beh
8. SSIS-P
Soc Skills
9. SSIS-P
Prob Beh

** p < .01; * p < .05

92

1

Table 13
Pearson Correlations Between Post-Test Measures
1.
1. Difficulty
Likert-A
2. Difficulty
Likert-P
3. SSIS-A
Soc Skills
4. SSIS-A
Prob Beh
5. SSIS-P
Soc Skills
6. SSIS-P
Prob Beh

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

1
.043

1

.122

-.220

1

-.147

.248

-.187

1

.255

-.553**

.405**

-.366*

1

-.217

.472**

-.324*

.479**

-.691**

1

7. ASSP

.093

-.661**

.112

-.183

.772**

-.520**

1

8. RADS-2

.005

.271

-.232

.700**

-.350*

.339*

-.312*

1

9. RCMAS-2

.172

.200

-.120

.704**

-.068

.145

-.042

.706**

1

10. IMI

-.012

.008

.329

-.085

.211

-.177

.243

-.020

-.008

1

.101

.114

-.131

-.040

-.213

.013

-.123

-.133

.596**

11. Attitude
-.140
Toward Tech
** p < .01; * p < .05

93

1

More specifically, given that social impairments are a primary symptom of ASD, the goal was to
explore whether or not social factors were predictive of functioning in academic or emotional
domains. Given the significant differences between adolescent and parent reporters on the SSIS
and the Likert scale ratings of skill difficulty, and given the sample size of the current study,
separate regressions were conducted to examine effects separately for adolescent and parent
reporters. Research question 4B (below) describes whether or not the differences between
reporters were significant.
All participant data at Time 1 were used to answer this research question (see Table 9
above for social skills means and standard deviations, and Table 14 below for social support,
depression, and anxiety).
Table 14
Means and Standard Deviations for Social Support, Depression, and Anxiety

n

Intervention Group
Time 1
Time 2
M
SD
M
SD

n

Control Group
Time 1
Time 2
M
SD
M
SD

RADS-2

19

47.95

8.15

47.49

10.14

23

54.04

9.34

51.96

9.55

RCMAS-2

18

48.42

7.28

46.58

8.53

23

53.22

11.02

50.43

11.25

MSPSS
Child

21

58.81

14.68

---

---

26

57.88

15.04

---

---

MSPSS
Parent

20

61.50

10.18

---

---

25

59.44

12.04

---

---

First, multiple regression was used to predict academic achievement, measured by GPA, from
adolescent SSIS scores, adolescent Likert ratings, and adolescent MSPSS scores (N=31). There
was independence of residuals, as assessed by a Durbin-Watson statistic of 1.532. The
assumptions of linearity, independence of errors, homoscedasticity, and normality of residuals

94

were met. Child SSIS scores and Likert ratings did not significantly predict academic
2

achievement as measured by GPA, F = 1.009, p = .404, R = .101. Though GPA is a more
commonly used measure of academic achievement, many participants did not report GPA and
therefore a regression was also run with grades as an outcome variable (N=47). Similarly, the
2

predictor variables also did not significantly predict school grades, F = .791, p = .506, R = .057.
Next, academic outcomes were explored with parent-reports of social skills, social
difficulty, and social support as predictors. A multiple regression was run to predict academic
achievement, measured by GPA, from parent SSIS scores, parent Likert ratings, and parents
MSPSS score (N=35). There was independence of residuals, as assessed by a Durbin-Watson
statistic of 1.729. The assumptions of linearity, independence of errors, homoscedasticity, and
normality of residuals were met. These variables significantly predicted GPA, F = 3.173, p =
2

.038, R = .235. Parent SSIS scores significantly added to the prediction (p < .05), whereas
parent Likert ratings and parent MSPSS scores did not. Regression coefficients and standard
errors can be found in Table 15 (below).
Table 15
Social Skills & GPA – Parent Ratings
Variable
Intercept
MSPSS
SSIS
Difficulty Likert

B
1.555
.008
.016
.001

SEB
.833
.009
.008
.013

β
.155
.390*
.018

Note. B = unstandardized regression coefficient; SEB = standard error of the coefficient; β =
standardized coefficient
* p < .05

95

As with the child data, a regression was also run with grades as the outcome variable. These
2

predictors also significantly predicted school grades, F = 3.940, p = .014, R = .216. As with
GPA, parent SSIS scores significantly added to the prediction (p < .05), whereas parent Likert
ratings and parent MSPSS scores did not. Regression coefficients and standard errors can be
found in Table 16 (below).
Table 16
Social Skills & School Grades – Parent Ratings
Variable
B
β
SEB
Intercept
1.793
1.034
MSPSS
.012
.010
.184
SSIS
.019
.009
.366*
Difficulty Likert
.004
.016
.040
Note. B = unstandardized regression coefficient; SEB = standard error of the coefficient; β =
standardized coefficient
* p < .05
Next, multiple regression was used to examine mental health outcomes at Time 1, with
adolescent report data on the SSIS, Likert rating, and MSPSS at Time 1 used as predictors
2

(N=44). These variables did not significantly predict depression, F = 1.183, p = .160, R = .120,
2

nor did they significantly predict anxiety, F = 2.245, p = .098, R = .144. Similarly, social
2

variables at Time 2 did not predict depression or anxiety at Time 2 (F = 1.073, p = .352, R =
2

.053; F = .980, p = .385, R = .049). Additionally, social variables at Time 1 also did not predict
2

2

depression or anxiety at Time 2 (F = .325, p = .808, R = .027; F = 1.471, p = .239, R = .112).
Finally, mental health outcomes at Time 1 were examined from parent report data on the
SSIS, Likert rating, and MSPSS scores at Time 1 (N=47). These variables significantly

96

2

predicted depression scores on the RADS, F = 3.293, p = .029, R = .187. Parent Likert ratings
significantly added to the prediction (p < .05), whereas Parent SSIS scores and parent MSPSS
scores did not. Regression coefficients and standard errors can be found in Table 17.
Table 17
Prediction of Time 1 Depression Scores From Time 1 Parent Social Skills Ratings
Variable
Intercept
MSPSS
SSIS
Difficulty Likert

B
49.532
-.071
-.073
.400

SEB
12.978
.126
.115
.200

β
-.088
-.115
.322*

Note. B = unstandardized regression coefficient; SEB = standard error of the coefficient; β =
standardized coefficient
* p < .05
2

Social variables at Time 2 did not predict depression at Time 2 (F = .2.872, p = .069, R = .131).
However, these variables at Time 1 did predict depression scores at Time 2 (F = 4.447, p = .009,
2

R = .270). See Table 18 below for regression coefficients and standard errors. None of the
variables significantly added to the prediction.
Table 18
Prediction of Time 2 Depression Scores From Time 1 Parent Social Skills Ratings
Variable
B
β
SEB
Intercept
50.395
14.939
MSPSS
.094
.145
.105
SSIS
-.229
.124
-.344
Difficulty Likert
.381
.226
.290
Note. B = unstandardized regression coefficient; SEB = standard error of the coefficient; β =
standardized coefficient
* p < .05

97

Similarly, these variables at Time 1 significantly predicted anxiety scores on the RCMAS at
2

2

Time 1 and Time 2 (F = 4.809, p = .006, R = .251; F = 4.398, p = .010, R = .268). Parent
MSPSS scores and parent Likert ratings significantly added to both predictions (p < .05),
whereas parent SSIS scores did not. Regression coefficients and standard errors can be found in
Tables 19 and 20 (below).
Table 19
Prediction of Time 1 Anxiety Scores From Time 1 Parent Social Skills Ratings
Variable
B
β
SEB
Intercept
18.394
13.470
MSPSS
.262
.131
.302*
SSIS
-.025
.119
-.037
Difficulty Likert
.623
.208
.465*
Note. B = unstandardized regression coefficient; SEB = standard error of the coefficient; β =
standardized coefficient
* p < .05
Table 20
Prediction of Time 2 Anxiety Scores From Time 1 Parent Social Skills Ratings
Variable
B
β
SEB
Intercept
23.390
15.486
MSPSS
.374
.151
.405*
SSIS
-.164
.129
-.238
Difficulty Likert
.498
.234
.366*
Note. B = unstandardized regression coefficient; SEB = standard error of the coefficient; β =
standardized coefficient
* p < .05
Similar to the outcomes for depression however, social variables at Time 2 did not predict
2

anxiety at Time 2 (F = .845, p = .437, R = .043).
Though not an initial goal of this study, problem behaviors were also examined as a
predictor of mental health outcomes due to the effects found in Research Question 1 and because

98

behavioral problems are another primary symptom of ASD. Linear regression was used to
predict RADS-2 scores and RCMAS-2 scores from SSIS Problem Behavior adolescent and
parent ratings. Adolescent Problem Behavior scores significantly predicted both depression (F =
2

2

15.822, p = < .001, R = .264) and anxiety (F = 33.232, p = < .001, R = .430) at Time 1 and at
2

2

Time 2 (F = 38.490, p = < .001, R = .490; F = 39.282, p = < .001, R = .495). In contrast,
2

parent Problem Behavior scores significantly predicted depression (F = 10.034, p = .003, R =
2

.182) at Time 1, but not anxiety (F = .514, p = .477, R = .011). These results held true at Time
2

2, with a significant prediction for depression (F = 5.076, p = .030, R = .115), but not anxiety
2

(F = .836, p = .366, R = .021).
Research Question 4B: Are these relationships different depending on adolescent- or
parent-report?
Given the differences in mean score differences between reporters, it was hypothesized
that the strength of the relationships between social, academic, and emotional variables may vary
as well. To test these differences, t-tests of regression weights from the regression models in
Research Question 4A were conducted. Each of the analyses indicated that the strengths of the
relationships between variables were not significantly different between reporters. T-scores and
significance levels can be found in the following Table 21 (below).

99

Table 21
Comparison of Parent-Adolescent Regression Weights
T-Score
1.103
.345
-1.351

.274
.731
.181

.921
-1.037
-1.166

.359
.302
.247

-.264
-.231
1.693

.792
.818
.094

-.190
1.114
.530

GPA
SSIS Reporter
MSPSS Reporter
Difficulty Likert Reporter
School Grades
SSIS Reporter
MSPSS Reporter
Difficulty Likert Reporter
RADS
SSIS Reporter
MSPSS Reporter
Difficulty Likert Reporter
RCMAS
SSIS Reporter
MSPSS Reporter
Difficulty Likert Reporter

p

.850
.268
.598

Research Question 5: Are adolescents’ self-reported changes in social skills related to
changes in mental health problems?
A final goal of this study was to examine whether or not changes in social skills from
Time 1 to Time 2 were predictive of any changes in mental health functioning (i.e. depression
and anxiety). To test this, multiple regression was used to predict changes in depression and
anxiety from changes in adolescent SSIS scores and adolescent Likert ratings (i.e., using gain
scores for each of the predictor and outcome variables; N=17). Prior to running the analysis, the
data were examined to ensure that assumptions of linearity, independence of errors,
homoscedasticity, unusual points and normality of residuals were met. Multiple regression
indicated that adolescent reported social skills changes significantly predicted changes in
2

anxiety, F = 4.747, p = .028, R = .422. Adolescent reported SSIS gains significantly added to

100

the prediction (p < .05), whereas adolescent reported Likert changes did not. Regression
coefficients and standard errors can be found in Table 22 (below).
Table 22
Social Skills and Anxiety – Adolescent Report
Variable
B
β
SEB
Intercept
-.253
1.386
SSIS Gain Score
-.403
.172
-.528*
Difficulty Likert Gain Score
.219
.211
.234
Note. B = unstandardized regression coefficient; SEB = standard error of the coefficient; β =
standardized coefficient
* p < .05
In contrast, social skills changes did not significantly predict depression changes, F = 2.484, p =
2

.122, R = .276. Adolescent-reported changes in problem behavior were not predictive of either
2

2

depression changes (F = 2.376, p = .131, R = .057) or anxiety changes (F = .854, p = .361, R =
.021).

101

CHAPTER 5
Discussion
Efficacy of the Online Intervention
One purpose of this study was to examine the effectiveness of an online social skills
intervention that utilized video modeling to teach social communication skills. Using a
randomized controlled trial design, participants were assigned to intervention and wait-list
control groups and pre/post measures were utilized to examine group differences. Results
indicated that there were no significant differences between the intervention and wait-list control
groups on social skills gains, as measured by any of the social skill measures for either reporter.
Conclusions. In contrast to this author’s hypothesis, the results of two separate
MANCOVA procedures indicated that based on both parent and adolescent report, social skills
scores at Time 2 were not significantly different between groups after controlling for scores at
Time 1. Due to the unexpected findings regarding social skills improvement, problem behaviors
were examined as well. A MANCOVA procedure examining the intervention effect on the
combined problem behaviors subscales revealed a significant difference between the two groups.
More specifically, the intervention group displayed fewer problem behaviors at Time 2 after
controlling for Time 1 scores. Similarly, an ANCOVA procedure examining the intervention
effect on the problem behaviors total score revealed a significant difference between groups.
From adolescent perspectives, there were no significant treatment effects, but the ANCOVA
results approached significance (p = .059) with the intervention group reporting lower total
problem behavior scores at Time 2.
Overall, these findings suggest that the intervention had an effect on reducing problem
behaviors, but did not increase social skills as intended. These findings raise a question of

102

whether measures of problem behavior might highlight some social progress prior to it being
apparent in social skill measures. Although the intervention group began with fewer reported
problem behaviors, their reported problem behaviors at Time 2 were significantly less than the
wait-list control group. This finding suggests that it may be worth considering other behavioral
outcomes of social skills interventions. Although the participants in this study may not have
mastered the selected skills, perhaps the intervention content provided them with the skills to
more appropriately regulate their behavior during difficult social interactions.
Additionally, this study highlights the importance of having a comparison group when
conducting intervention research. Both the intervention and wait-list control groups
demonstrated an increase in reported social skills between Time 1 and Time 2. Without a waitlist control group for comparison, the change over time could easily have been falsely attributed
to the intervention. Additionally, these study results highlight the importance of selecting
appropriate measures for evaluating intervention effectiveness. Although a comparison between
child and parent reports will be more clearly addressed below, it is important to highlight here
that measures did not reveal the same effects across parent and child forms. This highlights the
importance of selecting measures that are both sensitive to change and valid.
Possible explanatory factors. There are several possible explanations for these results:
nature/sensitivity of the instruments, insufficient dosage, weak treatment integrity, and
heterogeneity of sample. First and foremost, the intervention was targeted at six very specific
social communication skills. The SSIS and the ASSP are long survey measures that serve to
gather a broader assessment of social skills than may be appropriate for assessing change in this
particular assessment. There is emerging literature on the effectiveness of utilizing the SSIS for
progress monitoring, but it may not be appropriate as a measure for a 6-week online intervention.

103

Similarly, the ASSP has some support for progress monitoring, but may not be appropriate for
this particular intervention strategy either.
Measurement. When considering the differences in results found from the SSIS Social
Skills reports and the SSIS Problem Behavior reports, it is important to consider the items that
are included on the SSIS Social Skills and Problem Behavior scales and more importantly, to
remember that the SSIS is not a tool specifically for those with ASD. Many of the items on the
Social Skills scale survey more complex social skills than were taught in this intervention (e.g.,
advocating for oneself) or survey skills that are less observable (e.g. trying to understand
someone’s feelings). The intervention more specifically taught basic social behaviors, such as
how to enter a conversation (i.e. wait for a pause, comment on the current topic) and the
behaviors measures on the Problem Behavior scale may simply have been more sensitive to
change. As described in the literature review above, various methods have been used to measure
intervention effectiveness for those with ASD. The measures, as well as the study design (Wang,
Parrila, & Cui, 2012), can influence the conclusions drawn from study results. This finding
highlights the importance of clearly selecting a measure that is sensitive to the changes that one
expects to find. Direct observation, which has been used as the method for obtaining outcome
data in many social skills interventions and video modeling studies, may have proven to be more
sensitive to change in this study than the standardized rating scales chosen. Since the Problem
Behaviors scale more clearly surveys behaviors, this may explain why change was more clearly
observed on this scale that the Social Skills scale.
Likert scale ratings of difficulty may be most appropriate when assessing change in
particular skills, but using these scales may present additional challenges for those with ASD.
The scale used in this study was a 7-point scale with three anchors/descriptors (i.e., not at all

104

difficult, somewhat difficult, and very difficult). Although the reliability of this scale was strong
(.84 for adolescents, .81 for parents), it’s possible that a shorter scale, or one with more
anchors/descriptors would be a more valid measure for those with ASD. Individuals with ASD
often have difficulty with abstract thinking, and therefore Likert scales may need to be grounded
in clearer criteria.
Dosage and integrity. Additional explanations for the lack of group differences are
dosage and intervention integrity. Though the intervention was initially intended to be eight
weeks long, the intervention group received only six weeks of intervention due to the
technological challenges described in the Methods section. The amount of time during which
participants had opportunities to view the lessons and practice the skills may simply have been
too short. Additionally, intervention group participants completed an average of 4.5 lessons of
the possible six. The amount of content to which participants were exposed in those 4.5 lessons
may not have been quite enough to provide them with the requisite knowledge to perform better
in social situations. Finally, although there were suggestions for practice at the conclusion of
each weekly lesson, the nature of the online intervention did not allow for required, structured
practice opportunities. Efforts were made to promote a higher level of treatment integrity, but
there were limitations within the implementation of the intervention that may have influenced
participants’ overall exposure to and experience with the intervention (see Limitations below).
Overall, one might not expect four lessons with few practice attempts to lead to clear increases in
social skills compared to those who do not receive the intervention at all. Although it would be
ideal to examine effects for those who complete all six lessons compared to those who complete
zero, the sample size was simply not large enough to do so in the current study.
Sample heterogeneity. Additionally, it is important to address the heterogeneity of the

105

current sample and some of the group differences between the intervention and wait-list control
groups. The qualitative feedback provided after each lesson, though not utilized for statistical
analysis, was valuable in that in highlighted that some participants already knew the target skills.
For participants who already knew these more basic social communication skills, one would not
expect to see a change in social skills ratings over time for those participants. Also, many of the
above-described analyses revealed that the intervention group reported higher mean social skills
and fewer mean problem behaviors at the beginning of the intervention. Random assignment
was used to create these groups, but it may have possibly left less room for improvement within
the intervention group. Whereas some participants may have already possessed the targeted
skills, others may have had impairments in social functioning and challenges with social
cognition that interfered with their abilities to interpret the videos in a meaningful way.
Alternatively, some participants may have been able to watch the videos and answer the quiz
successfully by recognizing cues that were specifically taught, but may have been unable to
generalize their knowledge to novel social situations given their social cognition deficits.
In addition to the variability in skills between participants and across groups, there may
have been other participant factors that made this intervention less successful than anticipated.
Social learning theory suggests that it is important that individuals observe others who are
similar to them in order for the best learning outcomes to occur. The videos were made with this
theory in mind and a diverse group of students were selected to film the video clips with both
males and females acting out each target behavior. It is possible however, that the participants in
this study did not identify with the actors in the video and therefore did not benefit as much from
the video modeling intervention. This may have been particularly likely for participants who
already felt competent with the target skills. Similarly, sociocultural theories emphasize the

106

importance of individuals’ experiences and the specific examples used in the videos may not
have been relevant for the study participants’ lives.
Practical significance. It is important to consider the clinical significance of the above
findings as well. A significant treatment effect was not found on the SSIS, but examination of
the descriptive categories in which participants’ scores fell did indicate some improvement (see
Table 23). The intervention group had fewer participants in the Well Below Average Category
and more participants in the Average category by the end of intervention, whereas the wait-list
control group had more participants in the Well Below Average category and fewer participants
in the Below Average Category by the end of the six-week study. This finding, along with the
Problem Behaviors results, highlight the importance of carefully considering the most
appropriate way to measure progress when examining intervention effectiveness.
Table 23
Changes in Parent-Reported SSIS Descriptive Categories
Well Below
Average
Time 1
Time 2

Below Average
Time 1

Time 2

Average
Time 1

Time 2

Intervention

5

3

6

6

8

10

Wait-List

5

9

11

7

6

6

Note. Well Below Average scores fall below 70, Below Average scores fall between 70 and 85,
Average scores fall between 85 and 115 (M = 100, SD = 15).
Perspectives on Effectiveness
A second goal of this study was to examine the influence of the reporter on the
interpretation of study outcomes. Given the limited research on social skill development and
intervention effectiveness from adolescent perspectives, and given this author’s thesis findings
that highlighted differences in perspectives on social skill difficulty (Hayter [Kuehnel], 2009),

107

examining both adolescent and parent perspectives was an important part of this project.
To begin, differences in parent and adolescent reports of social skills were examined at
both Time 1 to ensure that the expected differences in report were apparent with the current
sample. As expected, comparisons of SSIS scores and Likert scores were significantly different
between parents and adolescents at Time 1.
It was hypothesized that there would be some improvement in social skills for the
intervention group, and additionally, that adolescents and their parents would have different
reports of adolescents’ social skills improvement from pre- to post-intervention. The results of
Research Question 1 indicated that there was no significant social skills improvement for the
intervention group in comparison to the wait-list group for either reporter. Further analysis of
the Problem Behaviors scale however, revealed that there was a significant reduction in parentreported problem behaviors for the intervention group compared to the wait-list. Additionally,
this effect approached significance based on adolescent report. Therefore, the comparison
between intervention group reporters was still conducted.
The MANOVA procedure indicated that adolescents did not report changes that were
statistically significantly different from their parents on the SSIS or on the Difficulty Likert
scale. Although neither test reached significance, the trends in SSIS and Difficulty Likert ratings
were in the expected direction. Descriptive statistics suggested that adolescents tended to have a
more positive view of their improvements compared to their parents. In part, the lack of
significant findings is expected given the little change in skills over time and the small sample
size. With greater skill improvement and a larger sample, a stronger analysis of differences in
reporting would be possible. On the SSIS Problem Behaviors scale, it is interesting that
perspectives vary significantly between reporters at both Time 1, but there is strong agreement

108

between reporters in the overall change from Time 1 to Time 2. There is no clear explanation for
this finding and further research is warranted.
Conclusions. In general, it appears that adolescents seem to have a more positive view of
their social competence and behavioral presentation when compared to their parents’ reports.
This supports the sociocultural theoretical rationale for exploring multiple perspectives within
the current study. Had only parent or adolescent report been measured, the conclusions drawn
from this study would have been limited. Additionally, this has important clinical and
educational implications when thinking about how to present interventions or therapeutic
activities to adolescents with ASD. It is possible that those who have rather positive views of
themselves may view a lesser need for services, or that others’ suggesting a need for services
could damage relationships, motivation, or self-esteem. Those possible outcomes were not
investigated within the current study, but the relevance of adolescents’ perspectives is
highlighted given its contrast with parent perspectives. Other researchers are encouraged to
consider the potential impact of highlighting areas of difficulty when adolescents may not view
the same challenges, as this could potentially pose difficulties for adolescents, as well as for their
relationships with their parents.
In general, the main treatment effects that were found were within the parent report data,
and more specifically, when parents were rating problem behaviors. There are a few possible
explanations for these findings. Regarding the strength of parent perspectives, parents may be
more sensitive to change than adolescents. Given the challenges with social cognition that occur
for adolescents with ASD, it is possible that they do not recognize subtle changes in social
behavior. Parents, who more likely have typical social cognition skills, may be able to recognize
small improvements in adolescents’ behaviors that adolescents themselves do not notice. The

109

results from Research Question 2 did not indicate that parents were more sensitive to change
over time, but it is important to remember that only data from the intervention group was used to
compare rates of change across reporters. It would be important to examine in future research
how perspectives on change over time compare between those who are receiving intervention
services and those who are not, to gather additional information regarding adolescents’ abilities
to accurately evaluate their skill acquisition.
Another possible explanation is that parent perspectives may have been biased given that
they were not blinded to group assignment. Regarding the strength of the problem behaviors
ratings, it’s possible that the negative behaviors are more salient to parents and adolescents.
Research has demonstrated that individuals’ memory for negative events tends to be more
detailed than memory for neutral or positive events (Kensinger, 2011). Some may consider
social skills and problem behaviors a continuum of social behaviors, with problem behaviors
having negative valence and social skills having positive valence. In that case, the literature on
recognition and memory for various events may provide some support for the findings within the
problem behavior data. Additionally, it’s possible that the nature of the intervention simply
helped adolescents learn what not to do, but did not provide appropriate instruction for
replacement behaviors. Alternatively, the incorrect behaviors may have been more compelling
to study participants than the correct examples. Kochanska’s work on production compliance
(i.e. “Do”) and inhibition (i.e. “Don’t”) has demonstrated that “Do contexts pose a greater
regulatory challenge to young children” (Kochanska, Coy, & Murray, 2001, p. 1093). While this
study examined adolescents’ behaviors and Kochanska’s work has focused on young children,
social development processes for those with ASD may be more similar to younger children than
typically developing teenagers. This is something that should be more closely examined within

110

future research.
Influence of Attitudes on Intervention Outcomes
In an effort to understand why the online intervention may be more effective for some
adolescents than others, a goal of this study was to examine the influence of attitudes on
intervention outcomes. In particular, it was thought that variables such as one’s perceived
competence for learning, one’s interest in the social skills intervention, and one’s attitudes about
the online system used for intervention might predict the likelihood of intervention effectiveness.
Using multiple regression, these variables were examined as predictors of social skills gain from
adolescent and parent perspectives, changes in reported difficulty from both adolescent and
parent perspectives, and changes in anxiety and depression from adolescent perspectives. None
of the regression models proved to be significant however, suggesting that attitudes are not
predictive of intervention outcomes for this particular sample. It is important to note, that the
sample size required to achieve adequate power for these regressions is much larger than the
sample from the intervention group in this study. Therefore, it would be appropriate to reexamine this question with a larger sample size in the future.
Relationships Between Social, Academic, and Emotional Variables
In addition to examining the possible effects of an online social skills intervention, a
secondary goal of this study was to explore the relationship between social variables and other
academic and emotional variables. Although social impairments are characteristic of those with
ASD, not all adolescents with ASD necessarily experience academic or mental health
impairments. As individuals with ASD navigate adolescence, their social world changes and
becomes more complicated. Struggling during this time may potentially have an influence on
other areas of functioning, but it is not well known how this may occur. Some prior research has

111

demonstrated correlations between social functioning, anxiety, and depression, but more clearly
understanding patterns in symptomatology may help providers know how to provide effective
services to those with ASD.
Regression models at Time 1. Multiple regression was used in this study to examine
whether or not social functioning was predictive of academic or mental health functioning based
on both adolescent and parent report. Parent reports of social variables significantly predicted all
four scores. More specifically, at Time 1, 23.5% of the variance in GPA, 21.6% of the variance
in grades, 18.7% of the variance in depression, and 25.1% of the variance in anxiety were
explained by parent ratings of social variables (i.e., SSIS Social Skills, MSPSS, and Difficulty
Likert). Overall, adolescent reports of social variables did not predict academic GPA, school
grades, anxiety, or depression. However, adolescent reports of problem behavior did predict
both anxiety and depression. More specifically, at Time 1, 26.4% of the variance in depression
and 43.0% of the variance in anxiety were explained by adolescent ratings of problem behaviors.
An examination of the difference in regression weights of each model indicated that there
were no significant differences between reporters. Once again however, the sample size was
much too small to run this analysis with confidence in the results given the lack of power.
Regression models of change over time. Finally, an important purpose of this study was
examining the influence of change in social skills on changes in mental health outcomes. Given
the hypothesis that social skills would be negatively correlated with mental health challenges at a
given point in time, it was also hypothesized that improvement in social skills may be predictive
of a decrease in mental health problems. Multiple regression indicated that adolescent-reported
social skill changes were in fact predictive of anxiety changes, but were not predictive of
depression changes. As with the other findings discussed above, the power for this analysis was

112

low and therefore further analyses with a larger sample are warranted.
Conclusions. Overall, it turns out that parent perceptions of social skills seem to be
better at predicting outcomes than the adolescents’. Parent reports of social skills at Time 1
explained 23.5% of the variance in GPA, 21.6% of the variance in grades, 18.7% of the variance
in depression, and 25.1% of the variance in anxiety at Time 1. However, adolescents’
perceptions of problem behaviors seem to be better at predicting mental health outcomes than
parents. Adolescent reports of problem behaviors at Time 1 explained 26.4% of the variance in
depression scores and 43.0% of the variance in anxiety scores. Prior research has shown that
those with ASD have higher symptoms of depression and anxiety than typical populations (see
e.g., Mazurek & Kanne, 2010; Russell & Sofronoff, 2005), but that those with fewer ASD
symptoms have higher levels of anxiety and depression (Mazurek & Kanne, 2010). Therefore, it
was expected that social skill ratings would be predictive of anxiety and depression. However,
the adolescent-report data may be more indicative of a measurement issue. It’s possible that
because adolescents with ASD lack social understanding, they are not reliable reporters of social
skills as measured on the SSIS. Perhaps, however, they more easily understand the behaviors
described on the SSIS Problem Behaviors scale and can better evaluate their own skills and
report changes on a measure that is more discrete. It would be important for clinicians and
researchers in the future to be aware that self-report social skills measures have this limitation.
Finally, it is important to consider the overall mental health quality of the current
participants. At Time 1, 74% of wait list participants and 100% of intervention group
participants had anxiety scores within the average range. Similarly, 78% of wait list participants
and 95% of intervention group participants had depression scores within the normal range.
There was little room for clinical improvement within the intervention group and more room for

113

improvement among the wait list control group. As stated above, with lower baseline social
skills, higher baseline problem behaviors, and higher baseline rates of anxiety and depression,
the wait list group may show a greater benefit from this intervention than seen within this study.
Limitations
The results and conclusions that can be drawn from this study are limited by several
methodological and practical constraints. First and foremost is the small sample size. A power
analysis, described in the Methods section, identified the need for a minimum of 128 adolescent
participants across the two groups. This sample size was not obtained due to various factors
beyond the control of the researcher. The actual number of participants enrolled in the study was
much lower than the original estimates given by the recruitment site. This could have been
addressed however, by continuing recruitment efforts. Unfortunately, the untimely closure of the
website hosting the intervention resulted in the early termination of the intervention and
excluded further recruitment. Thus, the resulting sample of 50 did not afford an adequate level of
power for many of the statistical analyses and these findings should be interpreted with caution.
An additional limitation of this study is the lack of face-to-face contact with participants
during the completion of study measures and during the intervention phase. A primary impetus
for this study was to reach adolescents who may not otherwise access services, but doing so in a
distant manner presents its own challenges. Many of the measures used in this study can be
complicated to complete and are typically explained by the administrator of those measures.
Every effort was made to ensure that participants had the resources to complete survey measures
accurately, but there are limitations to administering these forms by mail. Additionally, there is
no way to verify that the participant who claimed to fill out the measure was in fact the one
doing so. It’s also possible that, although instructed to complete the measures on their own,

114

participants may have had help from others to read the questions or to form answers. The
presence of another person may have introduced response bias and threats to validity. It is
unknown the extent to which the measures were completed as directed.
Similarly, although data were collected on the intervention dosage, there is no guarantee
that those who logged into the website and completed the lessons were in fact, the intended
participants and not a parent, sibling, or friend. With that said, each of the participants who
enrolled in the study did so because either they, their parents, or both parties wanted to
participate. Without firsthand measures of intervention integrity, it is difficult to know with any
certainty that participants completed each portion of the study as they agreed to.
Two of the steps in the evidence-based practice guidelines for video modeling (described
above) are identifying skill deficits and determining whether those challenges are skill or
performance deficits. One weakness in this study is that there was no assessment of skills prior
to enrollment or participation in the intervention. Although the goals of the intervention were
clear in the recruitment advertisement and participants knew what they were signing up for,
some of the qualitative feedback from adolescents suggested that the skills being taught were not
areas of need for this group (e.g., “I already knew how to do this.”). In the future, a screener
may be appropriate to determine whether or not participants are a good fit for the intervention.
Another limitation of this study is the lack of observable outcomes. Although the survey
measures provide some worthwhile information about outcomes, it would be more ideal to have
observable measure of social communicative behavior to more clearly indicate whether or not
adolescents could use the taught skills effectively.
The validity of adolescent self-report may seem limited for those with ASD, but a
sociocultural perspective, as well as some leading researchers in the field (e.g., Bellini), argue

115

that individuals’ perceptions and construction of meaning are critical. Parent report may also be
limited in identifying social skill deficits for adolescents, given the amount of time adolescents
spend away from their parents each day, but similarly, their perspective, however limited, is key.
Finally, sensitivity to change over an 8-week period is not something that all measures clearly
assess.
Finally, other study limitations fall within the limits of the Knoodle online system. Many
participants chose not to use the discussion forum, rate the weekly lessons, or provide feedback
on whether or not they had practiced the target skills. The reasons for their lack of response are
unknown. Although Knoodle did not have the capability to make these components required,
perhaps using another course management system with this feature would have ensured greater
engagement. Additionally, it would have been ideal to provide differentiated feedback to the
weekly quizzes, but this was not possible with Knoodle. Lastly, although the quality of most
video clips was strong, the video clips that were not recorded by the drama club were lower
quality than those recorded by the drama club. There is no published research (to this author’s
knowledge) that suggests video quality has a direct effect on the outcomes of video modeling,
but it is possible that the quality could affect engagement.
Future Research
First, this researcher plans to re-run these analyses with a larger sample after the wait list
group finishes the intervention. It will of course be important to check for differences between
the two groups based on the intervention websites, but if there are no significant differences
between groups combining the data sets will allow for a more robust analysis of intervention
effects. If there are differences between the two groups, this unexpected change in procedures
will have provided opportunities for a closer investigation of the necessary intervention

116

components (e.g., differential feedback). A preliminary analysis of descriptive data from the
wait list control group on the Mindflash site reveal a higher percentage of lessons completed, as
well as higher weekly ratings. The participants on the Mindflash site will have an opportunity to
complete all eight lessons, which will allow for an analysis of the intended treatment package.
Also, given that the wait list group had lower mean social skills and higher mean problem
behaviors at baseline, floor and ceiling effects can be more clearly ruled out. Additionally, the
qualitative feedback provided by participants through weekly lesson ratings can be examined to
gather further information about participant satisfaction and how to improve the lessons. These
data will all help to refine the intervention package for future research. An enhancement to
consider based on the results of the current study and some of the qualitative feedback received
from participants is to alter the format for selecting lesson content. Rather than a pre-packaged
intervention, a pre-test to generate a course of relevant lessons for individuals may prove more
successful.
To gather additional data on the effectiveness of this online intervention strategy, it
would be important for researchers to develop sustainable systems of delivery. The unexpected
closing of the Knoodle course management site posed a significant challenge to this research
project and using a stable delivery system would be wise in the future. Similarly, careful
consideration of measures that could be utilized online to accurately gather relevant data would
be of importance to future researchers. The distribution of study measures was challenging in
this project, and there was a 1-3 week lag time between survey completion and intervention
implementation at both Time 1 and 2. A set of useful measures that could be utilized online in
closer proximity to intervention implementation would be helpful for future research. Similarly,
a screener that identified initial levels of social skills (as described in the Limitations section)

117

could strengthen future research as well.
Given the discrepancies between reporters seen in this study, it would be important to
carefully examine the social skills measures used for progress monitoring in future research in
general, and for monitoring online instruction in particular. For example, Bellini & Hopf (2007)
suggest, “Future research should also examine the reliability of the ASSP across informants and
the concurrent validity of the instrument…The latter could be accomplished by comparing the
results of the ASSP with more established social skills measures, such as the Social Skills Rating
System. A moderate correlation between the ASSP and these other instruments would be
expected, but not a high correlation, as these instruments were designed for general populations
and the ASSP was designed specifically for children with ASD” (p. 86). This researcher was
able to examine the ASSP and the SSIS and though a there was a high correlation between
measures at Time 1 (r = .850, p = < .001) and at Time 2 (r = .772, p = < .001), the measures did
not reveal similar findings within each of the research questions discussed above. Therefore,
further examination of the appropriateness of each measure for particular clinical and research
purposes is warranted. Also, examination of teacher report, as compared to parent and child
report may yield interesting findings with relevant clinical implications.
Additionally, future research should explore opportunities for direct behavioral
observation in distance-based interventions. Though not possible for the current study, future
research may examine distance-technology that allows for participant interaction with the
researcher to demonstrate learned skills by either self-recording video or video conferences.
Additionally, the current study did not address maintenance of learned skills, but future research
may examine not only the immediate effects of online strategies, but also the long-term effects.
It is not only important to know if a student can learn a skill, but also if the behavior generalizes

118

and sustains.
Lastly, the introduction and literature review in this dissertation highlighted the need to
consider individual deficits, needs, and perspectives when planning interventions. Although the
pre-dissertation study was a sincere effort to select target intervention skills that would be of
relevance to adolescents with ASD, there may still have been a mismatch between the taught
skills and the needs of participants. In the future, researchers should consider how to best match
the content to the needs of participants in a controlled manner, prior to increasing the amount of
content.
It would be important to gather further information about the role of parents in future
research as well. This study did not examine the level of parent involvement, and it is quite
possible that involvement varied among participants. While some parents may have watched the
videos with their children and prompted practice attempts, other parents may have been quite
uninvolved. It would be important for future research to examine the importance of parent
involvement and to assess parent’s satisfaction with the intervention as well. While some
parents prefer to have an active role in intervention or treatment, others prefer to more quietly
support their child’s independence in treatment. These data would be important for determining
how to provide socially valid treatments that maintain effectiveness when more broadly
distributed.
Similarly, the role of the researcher is one that needs to be more clearly investigated in
future research. While the researcher did not directly interact with any of the participants in this
study, there were efforts made to provide feedback to participants through the quiz function.
Additionally, the researcher monitored the discussion board and was available to participants by
email and occasionally, instant message. These mechanisms were put in place so that

119

participants were not engaging in the intervention without some oversight from the researcher,
who had social skills training experience. Despite some of the positive effects found within this
study, the limitations discussed above suggest a need for more possible feedback opportunities
and perhaps greater interaction between the researcher and participants in order to assess
outcomes. Future researchers should carefully examine these factors prior to more widely
implementing online interventions that include little implementer oversight. Though online or
computer-mediated interventions have become quite popular and may have far-reaching positive
effects, it is still unclear what are the key components of such interventions and this study
highlights the need for further empirical researcher prior to any widespread implementation or
commercialization.
Implications for Practice
Researchers have suggested that if evidence can demonstrate that computer-based
delivery of video modeling is an instructionally sound practice, “the dissemination of quality
products to teachers and students becomes a real possibility” (Ayres & Langone, 2005).
Teachers play a critical role in providing services to students on a day-to-day basis and could be
an effective resource for naturalistic instruction and increasing generalization or practice
opportunities. The lack of significant treatment effects on social skills development within the
current study has important implications for practice, however. Rushing into widespread
implementation of this evidence-based strategy in new delivery formats would not be wise until
those delivery methods have been more fully tested with high treatment integrity and found to be
effective. Additionally, it is critical to clearly define the role and responsibilities of teachers or
other implementers and to understand how their involvement may influence outcomes. It is not
yet clear what the critical components of computer-based interventions are, or how the

120

interventionist influences outcomes, and therefore clear guidelines for implementing computerbased delivery of video modeling have not yet been established. Once the key components are
identified, finding ways to ensure treatment integrity would be important. Overall, educators
could play a key role in delivering computer-based interventions in the future, but many factors
need to be explored prior to widely disseminating any computer-based program.
The role of parents was addressed above for considerations in future clinical research, but
it is similarly important to consider the role of parents within clinical practice. This study
demonstrated that the input from parents is important, if not critical, in understanding the
adolescent experience of those with ASD. While parent report had more predictive power in
some instances, adolescent report was more informative in other instances. Together, both
perspectives help provide a more complete understanding of the child’s functioning, as well as
his interaction with those in his environment. Gathering comprehensive data from multiple
perspectives has relevance for intervention and treatment when considering how one’s family
can play a role in providing services. With this intervention, the researcher had little control over
whether or not participants chose to engage in practice, or generalization, opportunities. Parents,
however, could play a key part in providing those opportunities for adolescents. More generally,
if clinicians consider the ways in which families can participate in treatment and intervention
services, the opportunities for learning, exposure, and reinforcement could be increased.
Conclusions
Overall, this study examined the effects of an online social skills intervention with video
modeling from both parent and adolescent perspectives. Additionally, this study examined the
relationships between social skills, academic achievement, and mental health among adolescents
with ASD. Although there is substantial literature on the effectiveness of social skills

121

interventions, and video modeling in particular, there is little research that examines alternative
forms of delivery. Positive treatment effects were not observed on social skills ratings, but
improvements in problem behaviors were seen for the intervention group. This study certainly
does not refute the years of research demonstrating the effectiveness of social skills interventions
and video modeling strategies, but rather highlights implications for alternative delivery of such
interventions. Also, this study highlighted the importance of carefully choosing measures for
evaluating the effectiveness of interventions, as there were mixed results depending on
measurement tool and reporter. Finally, this study highlighted the relationships between social
skills, problem behaviors, and academic and mental health outcomes that should continue to be
examined from both adolescent and parent perspectives. It is important for clinicians and
researchers to continue to think about creative ways to serve those with ASD, so that all
individuals who desire services can access them, and so those that may be at risk for poor mental
health outcomes have access to services to remediate those risks.

122

APPENDICES

123

APPENDIX A

Recruitment Letter

124

Recruitment Letter

Dear IAN Research participant,
Based on your family profile, one or more members of your family may qualify for the study
below. You should click on the study link or contact the study team directly, using the
information provided, if you are interested in joining. You do not have to participate in this study
and your non-participation will neither affect the care you receive from any health provider nor
your standing as a participant in IAN Research.
Please note that IAN Research is serving as a resource linking the autism community and
researchers. This study is not endorsed by or performed under the auspices of the IAN Research
project at Kennedy Krieger Institute/Johns Hopkins.
______________________________________________________________________________
Name of Study:

An Online Social Skills Intervention with Video Modeling for
Adolescents with ASD
Institution:
Michigan State University, East Lansing, Michigan
Location:
Online study; no geographic limitation within the United States
Eligibility Criteria: Children diagnosed with an Autism Spectrum Disorder, in grades 9-12,
between 13-18 years of age, who (i) spend more than 50% of their time
in general education classes, (ii) have basic computer skills, (iii) have
regular Internet access, and (iv) read at or above a fifth grade reading
level.
Principal Investigator: Dr. Evelyn R. Oka, Associate Professor, School Psychology &
Educational Psychology
Research Contact: Carolyn Hayter, M.A., Doctoral Candidate in School Psychology
Email: adolescencestudy@gmail.com
Phone: 513-ASD-PROJ (513-273-7765)
Study Link:
https://docs.google.com/spreadsheet/viewform?formkey=dGZ0amptcmx3V3RBWW5jM2d0T0p
6Qnc6MQ#gid=0
Dear Parent,
I am writing this letter to offer your child an opportunity to participate in a research study
through Michigan State University. Many adolescents with autism spectrum disorder (ASD)
struggle with interpersonal communication and we are interested in understanding how to
effectively support the development of adolescents’ social communication skills. The study
examines the effectiveness of an online intervention that has been developed to teach social
communication skills to adolescents with ASD.
The intervention includes a series of fun videos, online presentations, brief quizzes, and surveys,
as well as opportunities to interact with other research participants in online discussions (i.e.,
other adolescents with ASD). The discussion forums will be carefully monitored throughout the
duration of the study to protect participants from possible inappropriate comments, bullying, etc.

125

Adolescents will be expected to participate in one online lesson per week for a total of eight
weeks. Each weekly lesson will last approximately 15-30 minutes. However, adolescents may
spend additional time on the intervention website if they choose to view lessons multiple times
or if they choose to communicate with other participants on a more frequent basis.
Once enrolled, families will be sent a packet by mail that includes consent forms and
questionnaires. Adolescents and one parent of each adolescent will be requested to each
complete multiple questionnaires. The questionnaires will take approximately 60 minutes total
for each adolescent and parent. More than one child per family may participate in the study, but
parents will need to fill out separate questionnaires for each child. When completed, signed
consent forms and questionnaires must be returned in a postage-paid envelope.
After consent forms and questionnaires are returned, adolescents will be randomly assigned to
two groups. One group will be asked to begin the intervention immediately and the other will be
put on a waiting list. Regardless of which group adolescents are in, all adolescents and one
parent of each adolescent will be expected to complete a second set of questionnaires after
approximately eight weeks. After the second set of questionnaires is returned, all adolescents on
the waiting list will begin the intervention. Approximately eight weeks later, a third and final set
of questionnaires will be sent to all participants.
Each time parents and adolescents both return completed forms, families will be provided with a
$10 gift card for Amazon.com. This will occur after each of the three rounds of questionnaires,
allowing for families to earn a total of $30 in Amazon.com gift cards.
Those who participate fully throughout the duration of the study (i.e., parents and adolescents
both complete and return all measures, adolescents participate in at least seven of eight
intervention lessons) will be entered in a raffle at the conclusion of the study for a pocket video
camera. One video camera will be raffled off for every 30 adolescent participants.
If you are interested in joining this study, please click
https://docs.google.com/spreadsheet/viewform?formkey=dGZ0amptcmx3V3RBWW5jM2d0T0p
6Qnc6MQ#gid=0
and complete the online enrollment form. If you have any questions about the study, please
contact Carolyn Hayter at 513-ASD-PROJ (513-273-7765) or adolescencestudy@gmail.com.
Thank you in advance for your time and assistance. We look forward working with you in the
near future.
Sincerely,
Carolyn Hayter, M.A.
Evelyn Oka, Ph.D.

126

APPENDIX B

Parent Informed Consent

127

Consent for Participation in Research
You and your child are invited to participate in a research study through Michigan State
University. Researchers are required to provide a consent form to inform you about the study, to
inform you that participation is voluntary, to explain risks and benefits of participation, and to
allow you to make an informed decision. The title of the study is An Online Social Skills
Intervention with Video Modeling for Adolescents with ASD. Michigan State University
graduate student, Carolyn Hayter, and Associate Professor, Dr. Evelyn Oka, are conducting the
research. You should feel free to ask the researchers any questions you may have.
PURPOSE OF RESEARCH:
Your child is being asked to participate in a research study on social communication. Your child
was selected as a possible participant because he/she has an autism spectrum disorder (ASD).
Many adolescents with ASD struggle with interpersonal communication and we are interested in
understanding how to effectively support the development of adolescents’ social communication
skills.
The study examines the effectiveness of an online intervention that has been developed to teach
social communication skills to adolescents with ASD. We are recruiting adolescents with ASD in
grades 9-12 who are 18 years of age or younger. To be included in the study, adolescents must
spend at least 50% of their time at school in general education classes, have basic computer
skills, have regular Internet access, and read at, or above, a fifth grade level.
WHAT YOU AND YOUR CHILD WILL DO:
Adolescents and one parent of each adolescent are asked to each complete five questionnaires at
the beginning of the study. These questionnaires will take about 60 minutes for each adolescent
and parent to complete. After consent forms and questionnaires are returned, adolescents will be
randomly assigned to two groups. One group will be asked to begin the intervention
immediately and the other will be put on a waiting list. You will receive an email that tells you
which group your child has been assigned to. Regardless of which group adolescents are in, all
adolescents and one parent of each adolescent will be expected to complete a second set of
questionnaires after approximately eight weeks. Parents will have four questionnaires to
complete and adolescents will have six questionnaires to complete at this time. After the second
set of questionnaires is returned, all adolescents on the waiting list will begin the intervention.
Approximately eight weeks later, a third and final set of questionnaires (i.e., four questionnaires
for parents and six for adolescents) will be sent to all participants.
The questionnaires that parents will complete ask about adolescents’ social skills and social
support. Additionally, parents will complete a brief demographic survey. The questionnaires
that adolescents will complete ask about their social skills, social support, attitudes toward
technology, and their feelings (i.e., relatedness, anxiety, and depression).
The intervention includes a series of fun videos, online presentations, brief quizzes, and surveys,
as well as opportunities to interact with other research participants (i.e., other adolescents with

128

ASD). Adolescents will be expected to participate in and complete 8 weekly lessons that each
lasts approximately 15-30 minutes. Until the study ends, adolescents may spend more time on
the intervention website if they choose to view lessons multiple times or if they choose to
communicate with other participants on a more frequent basis. After adolescents complete the
eight weekly lessons, access to the online intervention ends and parents and adolescents will
complete the final sets of questionnaires.
POTENTIAL BENEFITS AND RISKS:
This study will help us to learn how to support adolescents’ social and emotional development.
The views of parents and adolescents are important to help us understand more about the social
experience of adolescents with ASD, and your participation may help to improve the way in
which schools support students during this time. It is also possible that adolescent participants
will develop social skills over the course of intervention, as well as improve their feelings of
social support and relatedness.
Having your child spend time online on the Internet may cause concern. While it is important
that adolescents complete the lessons on their own, it is also important that parents are confident
and comfortable with the content of the lessons. Therefore, parents will be provided access to
the intervention website with limited rights. Parents will be able to view all intervention content,
but will not be able to read or participate in the discussion forums.
One possible risk of this study is that peer interaction during intervention could lead to negative
social or emotional outcomes if negative interactions, rude comments, or inappropriate remarks
occur in the online discussions. However, peer interaction may also lead to the development of
friendships or increased feelings of social support. A participant agreement form will be shared
with participants prior to the start of intervention that details the norms and expectations for
communicating in the discussion forums to minimize this risk. Participants must agree to abide
by the proposed expectations before beginning intervention. Additionally, the researchers will
monitor the discussion forums to ensure that participants are adhering to the stated expectations
and they will intervene if needed. Participants may also report directly to the researcher any
problems or inappropriate comments they encounter. These will be removed immediately.
A second possible risk is that participants may incorrectly use learned social skills in real-life
situations. At the conclusion of each lesson, suggestions for practice will provide appropriate
uses of the target skills in real-life contexts to minimize this risk. Additionally, you can support
your child’s participation by asking him/her about the intervention and encouraging his/her
efforts to practice the skills.
A third risk of this study is that it involves spending time to complete the questionnaires and
participate in intervention. The intervention has been designed with the perspectives and
preferences of adolescents with ASD in mind however, so we hope that this time will be
enjoyable and meaningful for participants.
Finally, both parents and adolescents may become more aware of adolescents’ skill deficits or
weaknesses, but also may become more aware of adolescents’ strengths. To minimize the risk of
adolescents developing negative feelings about their abilities, participants will be able to view

129

each lesson multiple times throughout the 8-week intervention and will be allowed to retake
quizzes in order to maximize opportunities for success.
PRIVACY AND CONFIDENTIALITY:
The data for this project will be kept confidential to the greatest extent allowable by law. Upon
enrollment, all participants will be assigned ID numbers, which will be used in place of names on
questionnaires and other hard copy documents. For online activities, participants will be
assigned usernames and will be strongly encouraged not to share their full names when
communicating with other participants. At the conclusion of intervention, any identifying
information in electronic data will be de-identified by replacing names with ID numbers.
Any hard copies of documents, forms, surveys, or rating scales will be stored in a locked filing
cabinet for approximately three years and then will be destroyed. Electronic information and data
from the online intervention will be stored on the investigators’ computers and will be passwordprotected. It will be stored until the completion of the study, or for approximately 3 years. Only
the researchers and the MSU Institutional Review Board will have access to the research data.
The results of this study may be published or presented at professional meetings, but the
identities of all research participants will remain confidential. It will not be possible for readers
to know who participated in the study. Only the researchers and the MSU IRB will know who
participates in the study. Results will be available to you upon request, will consist only of
group data, and will not include any identifying information.
The investigator has ethical and legal obligations to report suspected child abuse or neglect, and
to prevent your child from carrying out any threats to do serious harm to themselves or others. If
keeping information obtained in this study private would immediately put your child or someone
else in danger, the investigators would release that information to protect your child or another
person.
YOUR RIGHTS TO PARTICIPATE, SAY NO, OR WITHDRAW:
Your participation and your child’s participation in this study are entirely voluntary, which
means that you are free to choose whether or not you want to participate. You and your child are
free to withdraw your participation at any time without consequence. You may also choose not to
answer particular questions without consequence. Your non-participation will neither affect the
care you receive from any health provider nor your standing as a participant in IAN Research.
COSTS AND COMPENSATION FOR BEING IN THE STUDY:
There is no charge to participate in the study. However, it is possible that you may incur
additional charges on your Internet plan if your child spends a large amount of time on the
intervention site. The researchers are NOT able to provide compensation for Internet fees.

130

Each time parents and adolescents both return completed forms, families will be provided with a
$10 gift card for Amazon.com. This will occur after each of the three rounds of questionnaires,
allowing for families to earn a total of $30 in Amazon.com gift cards.
Those who participate fully throughout the duration of the study (i.e., parents and adolescents
both complete and return all measures, adolescents participate in at least seven of eight
intervention lessons) will be entered in a raffle at the conclusion of the study for a pocket video
camera. One video camera will be raffled off for every 30 adolescent participants.
CONTACT INFORMATION FOR QUESTIONS AND CONCERNS:
If you have concerns or questions about this study, such as scientific issues, how to do any part
of it, or to report an injury (i.e. physical, psychological, social, financial, or otherwise), please
contact the researchers: Carolyn Hayter, M.A., by email: adolescencestudy@gmail.com; phone:
513-ASD-PROJ (513-273-7765); or Evelyn Oka, Ph.D. by email: evoka@msu.edu; phone 517432-0843; regular mail: 439 Erikson Hall, 620 Farm Lane, East Lansing, MI, 48824; website:
http://www.educ.msu.edu/cepse/SchoolPsychology/faculty.asp.
If you have questions or concerns about your role and rights as a research participant, would like
to obtain information or offer input, or would like to register a complaint about this study, you
may contact, anonymously if you wish, the Michigan State University's Human Research
Protection Program at 517-355-2180, Fax 517-432-4503, or e-mail irb@msu.edu or regular mail
at 207 Olds Hall, 408 West Circle Drive, MSU, East Lansing, MI 48824.
DOCUMENTATION OF INFORMED CONSENT
If you agree to participate in the research study, fill in your name, and sign below.
¨ I agree to participate in this study and consent for my child,
___________________________________, to participate in this research study.

___________________________________
Signature

_____________________________
Date

You have been given a duplicate copy of this form to keep.

131

APPENDIX C

Adolescent Assent Form

132

Adolescent Agreement for Participation in Research
Study Title: An Online Social Skills Intervention with Video Modeling for Adolescents with
ASD
This is a research study and you are being invited to participate because you are an adolescent in
high school with an autism spectrum disorder (ASD). In this study, the researchers hope to learn
how to support the development of adolescents’ social communication skills.
What will happen if I take part in this study?
If you agree to be in this study, your participation will involve completing multiple
questionnaires and completing an 8-week online intervention. The intervention includes a series
of fun videos, online presentations, brief quizzes, and surveys, as well as opportunities to interact
with other research participants (i.e., other adolescents with ASD). You will be expected to
participate in eight weekly lessons that each lasts approximately 30 minutes. Until the study
ends, you may spend more time on the intervention website if you choose to view lessons
multiple times or if you choose to communicate with other participants on a more frequent basis.
After you complete the eight weekly lessons, access to the online intervention ends and you will
complete the final sets of questionnaires.
You will be asked to complete five questionnaires at the beginning of the study. These
questionnaires will take a total of about 60 minutes to fill out. After completing and returning
this form and the questionnaires, you will be randomly assigned to one of two groups. One
group will be asked to begin the intervention immediately and the other will be put on a waiting
list. Your parent will receive an email that states which group you have been assigned to.
Regardless of which group you are assigned to, you will be expected to complete a second set of
six questionnaires after approximately eight weeks. This set of questionnaires will also take
approximately 60 minutes. After the second set of questionnaires is returned, all adolescents on
the waiting list will begin the intervention. Then, approximately 8 weeks later a third set of six
questionnaires will be sent to all participants. This set of questionnaires will also take
approximately 60 minutes.
The questionnaires that you will complete ask about your feelings, social interactions, and school
experiences. The questionnaires are for research purposes only.
Are there any risks to my privacy or me?
Some of the questionnaires ask you to rate yourself and may make you feel uncomfortable.
Also, you may at times feel uncomfortable interacting with other participants online. You are
free to use the online discussion forums as much or as little as you would like. You may have
difficulty applying skills you learn online to real life. We will provide suggestions to help you
with this, and you will have opportunities to contact the researcher if you need to.
The data for this project will be kept confidential to the greatest extent allowable by law.
Confidentiality means privacy. In other words, information that you share through

133

questionnaires and online activities will not be shared with your parents or anyone else without
your permission. Only the researchers and the MSU Institutional Review Board will have access
to the research data.
Upon enrollment, you will be assigned an ID number, which will be used in place of your name
on questionnaires and other hard copy documents. For online activities, you will be assigned a
username. We strongly encourage you not to share your full name when communicating with
other participants.
In some instances, the researchers are required to break confidentiality and share information
about you with your parents or law enforcement agencies. The researchers are required to report
suspected child abuse or neglect, and to prevent you from carrying out any threats to do serious
harm to yourself or others. If keeping information confidential, or private, would immediately
put you or someone else in danger, the researchers would release that information.
Any hard copies of documents, forms, surveys, or rating scales will be stored in a locked filing
cabinet for approximately two years and then will be destroyed. Electronic information and data
from the online intervention will be stored on the investigators’ computers and will be passwordprotected. It will be stored until the completion of the study, or for approximately 2 years.
The results of this study may be published or presented at professional meetings, but the
identities of all research participants will remain confidential. In other words, it will not be
possible for readers to know who participated in the study. Only the researchers and the MSU
IRB will know who participates in the study. Results will be available to you upon request, will
consist only of group data, and will not include any identifying information.
Are there benefits?
You may learn some new social skills from participating in this intervention. This research
could also potentially benefit others with ASD in the future. Your views are important to help us
understand what it’s like to be a high school student with ASD today. Your participation may
help to improve the way in which schools support students with ASD during adolescence.
Can I say “No”?
You may choose to say “no”. Participation is voluntary, you may choose not to participate at all,
or you may refuse to participate in certain parts or answer certain questions. You may stop
participating at any time without consequence.
Choosing not to participate will involve no penalty or loss of benefits to you.
Are there any payments or costs?
It does not cost anything to participate in this study. If you and your parents complete and return
all required questionnaires at the beginning of the study, your family will be provided with a $10

134

gift card for Amazon.com. This will occur after each of the three rounds of questionnaires,
allowing for your family to earn a total of $30 in Amazon.com gift cards.
Also, if you complete and return all measures and participate in at least seven of eight
intervention lessons, you will be entered in a raffle at the end of the study for a pocket video
camera. One video camera will be raffled off for every 30 adolescent participants.
Who can answer my questions about the study?
If you have concerns or questions about this study, such as scientific issues, how to do any part
of it, or to report an injury (i.e. physical, psychological, social, financial, or otherwise), please
contact the researchers: Carolyn Hayter, M.A. by email: adolescencestudy@gmail.com; phone:
513-ASD-PROJ (513-273-7765); or Evelyn Oka, Ph.D., by email: evoka@msu.edu; phone: 517432-0843; regular mail: 439 Erikson Hall, 620 Farm Lane, East Lansing, MI, 48824; website:
http://www.educ.msu.edu/cepse/SchoolPsychology/faculty.asp.
If you have questions or concerns about your role and rights as a research participant, would like
to obtain information or offer input, or would like to register a complaint about this study, you
may contact, anonymously if you wish, the Michigan State University's Human Research
Protection Program at 517-355-2180, Fax 517-432-4503, or e-mail irb@msu.edu or regular mail
at 207 Olds Hall, 408 West Circle Drive, MSU, East Lansing, MI 48824.
Documentation of Agreement to Participate in Research
If you agree to participate in the research study, fill in your name, and sign below.
¨ I, ____________________________________, agree to participate in this research study.

___________________________________
Signature

_____________________________
Date

You have been given a duplicate copy of this form to keep.

135

APPENDIX D

Demographic Questionnaire

136

Demographic Questionnaire
Person Completing Form:
¨ Mother
¨ Father
¨ Grandmother
¨ Grandfather
¨ Other (please specify): ________________________________________
Child’s Birthdate (Month, Day, Year): ________________________________________
Child’s Grade:
¨ 9th Grade
¨ 10th Grade
¨ 11th Grade
¨ 12th Grade
Child’s Gender:
¨ Male
¨ Female
Child’s Race/Ethnicity:
¨ Hispanic
¨ American Indian or Alaska Native
¨ Asian
¨ Black or African American
¨ Native Hawaiian or Other Pacific Islander
¨ White
Does your child have Autism Spectrum Disorder?
o Yes
o No
At what age was your child diagnosed: _______________________________________
Which of the following most accurately describes your child? Please check only one.
o Autism
o Asperger’s Disorder
o Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS)
o Childhood Disintegrative Disorder
o Rett’s Disorder
o None

137

Do any of these characteristics describe your child? Check all that apply.
o Talented & Gifted
o ADHD - Attention Deficit Hyperactivity Disorder
o LD - Learning Disability
o Please specify subject area: ________________________________________
o Physical Disability
o Please specify: ________________________________________
o Emotional Disorder
o Please specify: ________________________________________
o

Other: ________________________________________

Which of the following best describes your child’s class schedule at school? Please choose
one.
o He/she takes all Special Education classes.
o He/she takes all General Education classes.
o He/she takes mostly Special Education classes with a few General Education classes.
o He/she takes mostly General Education classes with a few Special Education classes.
o He/she is homeschooled.
Has your child ever been retained/held back in school?
¨ No
¨ Yes
o If yes, during what grade(s): ______________________
Has your child ever been promoted early in school?
¨ No
¨ Yes
o If yes, during what grade(s): ______________________
Currently, what are your child’s typical grades at school? Check all that apply.
¨ As
¨ Bs
¨ Cs
¨ Ds
¨ Fs
What is your child’s cumulative GPA:
_____________

138

Is your child currently receiving another social skills intervention?
¨ Yes
¨ No
If yes, please describe the intervention (i.e., what activities does he/she do, for how long,
and how often).
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
Has your child received social skills intervention in the past?
¨ Yes
¨ No
If yes, please describe the intervention and when it occurred (i.e., what activities does
he/she do, when did intervention occur, for how long, and how often).
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________

139

APPENDIX E

Parent Likert Scale Form

140

Please rate how important you think the following skills are for your child.
1
not
at all
important
1.
2.
3.
4.
5.
6.
7.
8.

2

3

Entering a Conversation
Exiting a Conversation
Two-Way Conversations
Beginning Get-Togethers
Ending Get-Togethers
Handling Disagreements
Handling Teasing
Asking for Help

4
somewhat
important
1
1
1
1
1
1
1
1

2
2
2
2
2
2
2
2

3
3
3
3
3
3
3
3

5

4
4
4
4
4
4
4
4

6

5
5
5
5
5
5
5
5

6
6
6
6
6
6
6
6

7
very
important
7
7
7
7
7
7
7
7

Please rate how difficult you think the following skills are for your child.
1
not
at all
difficult
1.
2.
3.
4.
5.
6.
7.
8.

2

Entering a Conversation
Exiting a Conversation
Two-Way Conversations
Beginning Get-Togethers
Ending Get-Togethers
Handling Disagreements
Handling Teasing
Asking for Help

3

4
somewhat
difficult
1
1
1
1
1
1
1
1

2
2
2
2
2
2
2
2

3
3
3
3
3
3
3
3

141

5

4
4
4
4
4
4
4
4

6

5
5
5
5
5
5
5
5

6
6
6
6
6
6
6
6

7
very
difficult
7
7
7
7
7
7
7
7

APPENDIX F

Adolescent Likert Scale Form

142

Please rate how important you think the following skills are.
1
not
at all
important
1.
2.
3.
4.
5.
6.
7.
8.

2

3

Entering a Conversation
Exiting a Conversation
Two-Way Conversations
Beginning Get-Togethers
Ending Get-Togethers
Handling Disagreements
Handling Teasing
Asking for Help

4
somewhat
important
1
1
1
1
1
1
1
1

2
2
2
2
2
2
2
2

3
3
3
3
3
3
3
3

5

4
4
4
4
4
4
4
4

6

5
5
5
5
5
5
5
5

6
6
6
6
6
6
6
6

7
very
important
7
7
7
7
7
7
7
7

Please rate how difficult you think the following skills are.
1
not
at all
difficult
1.
2.
3.
4.
5.
6.
7.
8.

2

Entering a Conversation
Exiting a Conversation
Two-Way Conversations
Beginning Get-Togethers
Ending Get-Togethers
Handling Disagreements
Handling Teasing
Asking for Help

3

4
somewhat
difficult
1
1
1
1
1
1
1
1

2
2
2
2
2
2
2
2

3
3
3
3
3
3
3
3

143

5

4
4
4
4
4
4
4
4

6

5
5
5
5
5
5
5
5

6
6
6
6
6
6
6
6

7
very
difficult
7
7
7
7
7
7
7
7

APPENDIX G

Multidimensional Scale of Perceived Social Support Parent Form

144

Instructions: We are interested in how you feel about the following statements. Read each
statement carefully. Indicate how you feel about each statement.
Circle the “1” if you Very Strongly Disagree
Circle the “2” if you Strongly Disagree
Circle the “3” if you Mildly Disagree
Circle the “4” if you are Neutral
Circle the “5” if you Mildly Agree
Circle the “6” if you Strongly Agree
Circle the “7” if you Very Strongly Agree
1. There is a special person who is around
when my child is in need.
2. There is a special person with whom my
child shares his/her joys and sorrows.
3. My child’s family really tries to help
him/her.
4. My child gets the emotional help and
support he/she needs from family.
5. My child has a special person who is a
real source of comfort to him/her.
6. My child’s friends really try to help
him/her.
7. My child can count on his/her friends
when things go wrong.
8. My child can talk about his/her problems
with family.
9. My child has friends with whom he/she
can share his/her joys and sorrows.
10. There is a special person in my child’s
life who cares about his/her feelings.
11. My child’s family is willing to help
him/her make decisions.
12. My child can talk about his/her problems
with his/her friends.

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

This parent report form was adapted from the Multidimensional Scale of Perceived Social
Support (Zimet, Dahlem, Zimet, & Farley, 1988).

145

APPENDIX H

Multidimensional Scale of Perceived Social Support Adolescent Form

146

Instructions: We are interested in how you feel about the following statements. Read each
statement carefully. Indicate how you feel about each statement.
Circle the “1” if you Very Strongly Disagree
Circle the “2” if you Strongly Disagree
Circle the “3” if you Mildly Disagree
Circle the “4” if you are Neutral
Circle the “5” if you Mildly Agree
Circle the “6” if you Strongly Agree
Circle the “7” if you Very Strongly Agree
1. There is a special person who is around
when I am in need.
2. There is a special person with whom I
can share my joys and sorrows.
3. My family really tries to help me.

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

4. I get the emotional help and support I
need from my family.
5. I have a special person who is a real
source of comfort to me.
6. My friends really try to help me.

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

7. I can count on my friends when things
go wrong.
8. I can talk about my problems with my
family.
9. I have friends with whom I can share my
joys and sorrows.
10. There is a special person in my life who
cares about my feelings.
11. My family is willing to help me make
decisions.
12. I can talk about my problems with my
friends.

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

1

2

3

4

5

6

7

This adolescent report form is the Multidimensional Scale of Perceived Social Support (Zimet,
Dahlem, Zimet, & Farley, 1988).

147

APPENDIX I

Intrinsic Motivation Inventory Perceived Competence for Learning

148

Please rate the following statements.
1
not
at all
true
1. I feel confident in my ability to 1
learn how to talk with peers.
2. I am capable of learning how to 1
talk with peers.
3. I am able to achieve my goals in 1
learning to talk with peers.
4. I feel able to meet the challenge 1
of learning how to talk with peers.

2

3

2

3

4

2

3

2
2

7
very
true

5

6

7

4

5

6

7

4

5

6

7

3

149

6

3

This form was adapted from the IMI (Ryan, 1982).

4
5
somewhat
true

4

5

6

7

APPENDIX J

Intrinsic Motivation Inventory Interest/Enjoyment

150

Please rate the following statements.
1
not
at all
true

2

3

1. I enjoyed doing this study very much.

1

2

3

4

2. This study was fun to do.

1

2

3

3. I thought this was a boring study.

1

2

4. This study did not hold my attention
1
at all.
5. I would describe this study as very
1
interesting.
6. I thought this study was quite enjoyable. 1

5

6

7

4

5

6

7

3

4

5

6

7

3

4

5

6

7

3

4

5

6

7

2

7. While I was doing this study, I was
thinking about how much I enjoyed it.

3

4

5

6

7

2

151

7
very
true

2

This form was adapted from the IMI (Ryan, 1982).

6

2

1

4
5
somewhat
true

3

4

5

6

7

APPENDIX K

Attitude Toward Using Knoodle

152

Please rate the following statements.
1
not
at all
true

2

3

4
5
somewhat
true

6

7
very
true

1. Using the online system was a bad idea. 1

2

3

4

5

6

7

2. The online system makes work more
interesting.
3. Working with the online system is fun.

1

2

3

4

5

6

7

1

2

3

4

5

6

7

4. I like working with the online system.

1

2

3

4

5

6

7

This form was adapted from the Attitude Toward Using Technology Scale (Venkatesh, Morris,
Davis, & Davis, 2003).

153

APPENDIX L

Diagnostic Criteria

154

Table 24
Diagnostic Criteria for Autistic Disorder
A. A total of six (or more) items from (1), (2), and (3), with at least two from (1), and one
each from (2) and (3):
1. qualitative impairment in social interaction, as manifested by at least two of the
following:
a. marked impairment in the use of multiple nonverbal behaviors such as
eye-to-eye gaze, facial expression, body postures, and gestures to regulate
social interaction
b. failure to develop peer relationships appropriate to developmental level
c. a lack of spontaneous seeking to share enjoyment, interests, or
achievements with other people (e.g., by a lack of showing, bringing, or
pointing out objects of interest)
d. lack of social or emotional reciprocity
2. qualitative impairments in communication as manifested by at least one of the
following:
a. delay in, or total lack of, the development of spoken language (not
accompanied by an attempt to compensate through alternative modes of
communication such as gesture or mime)
b. in individuals with adequate speech, marked impairment in the ability to
initiate or sustain a conversation with others
c. stereotyped and repetitive use of language or idiosyncratic language
d. lack of varied, spontaneous make-believe play or social imitative play
appropriate to developmental level
3. restricted repetitive and stereotyped patterns of behavior, interests, and activities,
as manifested by at least one of the following:
a. encompassing preoccupation with one or more stereotyped and restricted
patterns of interest that is abnormal either in intensity or focus
b. apparently inflexible adherence to specific, nonfunctional routines or
rituals
c. stereotyped and repetitive motor mannerisms (e.g., hand or finger flapping
or twisting, or complex whole-body movements)
d. persistent preoccupation with parts of objects
B. Delays or abnormal functioning in at least one of the following areas, with onset prior to
age 3 years: (1) social interaction, (2) language as used in social communication, or (3)
symbolic or imaginative play.
C. The disturbance is not better accounted for by Rett's Disorder or Childhood
Disintegrative Disorder.
Note. From DSM-IV-TR (American Psychiatric Association, 2000, p. 75).

155

Table 25
Diagnostic Criteria for Rett's Disorder
A. All of the following:
1. apparently normal prenatal and perinatal development
2. apparently normal psychomotor development through the first 5 months after
birth
3. normal head circumference at birth
B. Onset of all of the following after the period of normal development:
1. deceleration of head growth between ages 5 and 48 months
2. loss of previously acquired purposeful hand skills between ages 5 and 30 months
with the subsequent development of stereotyped hand movements (e.g., handwringing or hand washing)
3. loss of social engagement early in the course (although often social interaction
develops later)
4. appearance of poorly coordinated gait or trunk movements
5. severely impaired expressive and receptive language development with severe
psychomotor retardation
Note. From DSM-IV-TR (American Psychiatric Association, 2000, p. 77).

156

Table 26
Diagnostic Criteria for Childhood Disintegrative Disorder
A. Apparently normal development for at least the first 2 years after birth as manifested by
the presence of age-appropriate verbal and nonverbal communication, social
relationships, play, and adaptive behavior.
B. Clinically significant loss of previously acquired skills (before age 10 years) in at least
two of the following areas:
1. expressive or receptive language
2. social skills or adaptive behavior
3. bowel or bladder control
4. play
5. motor skills
C. Abnormalities of functioning in at least two of the following areas:
1. qualitative impairment in social interaction (e.g., impairment in nonverbal
behaviors, failure to develop peer relationships, lack of social or emotional
reciprocity)
2. qualitative impairments in communication (e.g., delay or lack of spoken language,
inability to initiate or sustain a conversation, stereotyped and repetitive use of
language, lack of varied make-believe play)
3. restricted, repetitive, and stereotyped patterns of behavior, interests, and activities,
including motor stereotypies and mannerisms
D. The disturbance is not better accounted for by another specific Pervasive Developmental
Disorder or by Schizophrenia.
Note. From DSM-IV-TR (American Psychiatric Association, 2000, p. 79).

157

Table 27
Diagnostic Criteria for Asperger's Disorder
A. Qualitative impairment in social interaction, as manifested by at least two of the
following:
1. marked impairment in the use of multiple nonverbal behaviors such as eye-to-eye
gaze, facial expression, body postures, and gestures to regulate social interaction
2. failure to develop peer relationships appropriate to developmental level
3. a lack of spontaneous seeking to share enjoyment, interests, or achievements with
other people (e.g., by a lack of showing, bringing, or pointing out objects of
interest to other people)
4. lack of social or emotional reciprocity
B. Restricted repetitive and stereotyped patterns of behavior, interests, and activities, as
manifested by at least one of the following:
1. encompassing preoccupation with one or more stereotyped and restricted patterns
of interest that is abnormal either in intensity or focus
2. apparently inflexible adherence to specific, nonfunctional routines or rituals
3. stereotyped and repetitive motor mannerisms (e.g., hand or finger flapping or
twisting, or complex whole-body movements)
4. persistent preoccupation with parts of objects
C. The disturbance causes clinically significant impairment in social, occupational, or other
important areas of functioning.
D. There is no clinically significant general delay in language (e.g., single words used by
age 2 years, communicative phrases used by age 3 years).
E. There is no clinically significant delay in cognitive development or in the development of
age-appropriate self-help skills, adaptive behavior (other than in social interaction), and
curiosity about the environment in childhood.
F. Criteria are not met for another specific Pervasive Developmental Disorder or
Schizophrenia.
Note. From DSM-IV-TR (American Psychiatric Association, 2000, p. 84).

158

REFERENCES

159

REFERENCES
Achenbach, T. M., & Rescorla, L. (2001). Manual for the ASEBA school-age forms & profiles:
An integrated system of multi- informant assessment. Burlington, VT: University of
Vermont, Research Center for Children, Youth & Families.
Adelman, H. S., & Taylor, L. (2000). Promoting mental health in schools in the midst of school
reform. Journal of School Health, 70, 171–178. doi:10.1111/j.1746-1561.2000.tb06467.x
Allan, S., & Gilbert, P. (1995). A social comparison scale: Psychometric properties and
relationship to psychopathology. Personality and Individual Differences, 19, 293–299.
Retrieved from http://www.journals.elsevier.com/personality-and-individual-differences/
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders
th
(4 ed., text rev.). Washington, DC: American Psychiatric Association.
doi:10.1176/appi.books.9780890423349
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental
disorders: DSM-V. Washington, DC: American Psychiatric Association.
Attwood, T. (1998). Asperger’s syndrome: A guide for parents and professionals. London, UK:
Jessica Kingsley Publishers Ltd.
Autism Speaks. (2010). Autism speaks. It’s time to listen. Retrieved from
http://www.autismspeaks.org/.
Ayres, K. M., & Langone, J. (2005). Intervention and instruction with video for students with
autism: A review of the literature. Education and Training in Autism and Developmental
Disabilities, 40, 183–196. Retrieved from
http://daddcec.org/Publications/ETADDJournal.aspx
Bandura, A. (1971). Social learning theory. New York: General Learning Press.
Barnhill, G. P. (2001). Social attributions and depression in adolescents with Asperger
syndrome. Focus on Autism and Other Developmental Disabilities. 16, 46-53.
doi:10.1177/108835760101600112
Barnhill, G. P., Cook, K., Tebbenkamp, K., & Smith Myles, B. (2002). The effectiveness of
social skills intervention targeting nonverbal communication for adolescents with
Asperger Syndrome and related pervasive developmental delays. Focus on Autism and
Other Developmental Disabilities, 17, 112-118. doi:10.1177/10883576020170020601
Barnhill, G. P., Hagiwara, T., Smith Myles, B., & Simpson, R. L. (2000a). Asperger Syndrome:
A study of the cognitive profiles of 37 children and adolescents. Focus on Autism and
Other Developmental Disabilities, 15, 146-153. doi:10.1177/108835760001500303

160

Barnhill, G. P., Hagiwara, T., Smith Myles, B., Simpson, R. L., Brick, M. L., & Griswold, D. E.
(2000b). Parent, teacher, and self-report of problem and adaptive behaviors in children
and adolescents with Asperger Syndrome. Assessment for Effective Intervention, 25, 147167. doi:10.1177/073724770002500205
Bauminger, N. (2002). The facilitation of social-emotional understanding and social interaction
in high-functioning children with autism: Intervention outcomes. Journal of Autism and
Developmental Disorders, 32, 283-298. doi:10.1023/A:1016378718278
Beaumont, R., & Sofronoff, K. (2008). A multi-component social skills intervention for children
with Asperger syndrome: The Junior Detective Training Program. Journal of Child
Psychology and Psychiatry, 49, 743-753. doi:10.1111/j.1469-7610.2008.01920.x
Bellini, S. (2006). Building Social Relationships. Shawnee Mission, KS: Autism Asperger
Publishing Co.
Bellini, S., & Akullian, J. (2007). A meta-analysis of video modeling and video self-modeling
interventions for children and adolescents with autism spectrum disorders. Exceptional
Children, 73, 264-287. Retrieved from https://www.cec.sped.org/Publications/CECJournals/Exceptional-Children?sc_lang=en
Bellini, S., & Hopf, A. (2007). The development of the Autism Social Skills Profile: A
preliminary analysis of psychometric properties. Focus on Autism and Other
Developmental Disabilities, 22, 80-87. doi:10.1177/10883576070220020801
Bellini, S., Peters, J. K., Benner, L., & Hopf, A. (2007). A meta-analysis of school-based social
skills interventions for children with autism spectrum disorders. Remedial and Special
Education, 28, 153-162. doi:10.1177/07419325070280030401
Benford, P. (2008). The use of internet-based communication by people with autism
(unpublished thesis). University of Nottingham, England.
Bernad-Ripoll, S. (2007). Using a self-as-model video combined with social stories to help a
child with Asperger Syndrome understand emotions. Focus on Autism and Other
Developmental Disabilities, 22, 100-106. doi:10.1177/10883576070220020101
Bishop, J. (2003). The Internet for educating individuals with social impairments. Journal of
Computer Assisted Learning, 19, 546-556. doi:10.1046/j.0266-4909.2003.00057.x
Blume, H. (1997, June 30). Autistics, freed from face-to-face encounters, are communicating in
cyberspace. The New York Times. Retrieved from
http://www.nytimes.com/1997/06/30/business/autistics-freed-from-face-to-faceencounters-are-communicating-in-cyberspace.html?pagewanted=print&src=pm

161

Bronfenbrenner, U., & Ceci, S. J. (1994). Nature-nurture reconceptualized in developmental
perspective: A bioecological model. Psychology Review, 101, 568-596. Retrieved from
http://www.apa.org/pubs/journals/rev/index.aspx
Burke, M., Kraut, R., & Williams, D. (2010, February). Proceedings from The 2010 ACM
Conference on Computer Supported Cooperative Work. Savannah, GA: ACM
Canty-Mitchell, J., & Zimet, G. D. (2000). Psychometric properties of the Multidimensional
Scale of Perceived Social Support in urban adolescents. American Journal of Community
Psychology, 28, 391-400. doi:10.1023/A:1005109522457
Center for Disease Control, Department of Health and Human Services. (2008, January 30).
Prevalence of Autism Spectrum Disorders - Autism and Developmental Disabilities
Retrieved from http://www.cdc.gov/ncbdd/autism/faq_prevalence.htm
Center for Disease Control, Department of Health and Human Services. (2010, May 13). Autism
Spectrum Disorders (ASDs): Data and Statistics. Retrieved from
http://www.cdc.gov/ncbddd/autism/data.html
Center for Disease Control, Department of Health and Human Services. (2012, March 30).
Autism Spectrum Disorders (ASDs): Data and Statistics. Retrieved from
http://www.cdc.gov/ncbddd/autism/data.html
Chambless, D. L., & Hollon, S. D. (1998). Defining empirically supported therapies. Journal of
Consulting and Clinical Psychology, 66, 7-18. Retrieved from
http://www.apa.org/pubs/journals/ccp/index.aspx
Cheng, S., & Chan, A. (2004). The Multidimensional Scale of Perceived Social Support:
Dimensionality and age and gender differences in adolescents. Personality and Individual
Differences, 37, 1359-1369. doi:10.1016/j.paid.2004.01.006
Church, C., Alisanski, S., & Amanullah, S. (2000). The social, behavioral, and academic
experiences of children with Asperger Syndrome. Focus on Autism and Other
Developmental Disabilities, 15, 12-20. doi:10.1177/108835760001500102
Dagnan, D., & Sandhu, S. (1999) Social comparison, self-esteem and depression in people with
intellectual disability. Journal of Intellectual Disability Research, 43, 372–379.
doi:10.1046/j.1365-2788.1999.043005372
Daniels, A. M., Rosenber, R.E., Anderson, C., Law, J. K., Marvin, A. M., & Law, P. A. (2012).
Verification of parent-report of child autism spectrum disorder diagnosis to a web-based
autism registry. Journal of Autism and Developmental Disorders, 42, 257-265.
doi:10.1007/s10803-011-1236-7
Daniels, H. (2005). An Introduction to Vygotsky. New York, NY: Routledge.

162

Daniels, H., Cole, M., & Wertsch, J. (2007). The Cambridge Companion to Vygotsky.
Cambridge, UK: Cambridge University Press.
Data Accountability Center, Individuals with Disabilities Education Act. (2011, November 11).
Part B Child Count. Retrieved from https://www.ideadata.org/partBchildcount.asp
Deardorff, J. (2008, April 2). Autistics don’t want to be cured. Chicago Tribune. Retrieved from
http://www.chicagotribune.com/
Derogatis, L. R., Lipman, R. S., Rickels, K., Uhlenhuth, E. H., & Covi, L. (1974). The Hopkins
Symptom Checklist (HSCL): A self-report symptom inventory. Behavioral Science, 19,
1-15. doi:10.1002/bs.3830190102
DiMaggio, P., Hargittai, E., Neuman, W., & Robinson, J. P. (2001). Social implications of the
Internet. Annual Review of Sociology, 27, 307-336. Retrieved from
http://www.jstor.org/stable/2678624
Durkin, K., Whitehouse, A., Jaquet, E., Ziatas, K., & Walker, A. J. (2010). Cell phone use by
adolescents with Asperger Syndrome. Research in Autism Spectrum Disorders, 4, 314318. doi:10.1016/j.rasd.2009.09.017
Durlak, J. A., Weissberg, R. P., Dymnicki, A. B., Taylor, R. D., & Schellinger, K. B. (2011). The
impact of enhancing students’ social and emotional learning: A meta-analysis of schoolbased universal interventions. Child Development, 82, 405-432. doi:10.1111/j.14678624.2010.01564.x
Egan, G. J. (1989). Assessment of emotional processing in right and left hemisphere stroke
patients: A validation study of the Perception of Emotion Test. (Unpublished doctoral
dissertation). Georgia State University, Atlanta, GA.
Eysenbach, G., Powell, J., Englesakis, M., Rizo, C., & Stern, A. (2004). Health related virtual
communities and electronic support groups: Systematic review of the effects of online
peer to peer interactions. British Medical Journal, 328, 1166-1170. doi:
10.1136/bmj.328.7449.1166
Farrugia, S., & Hudson, J. (2006). Anxiety in adolescents with Asperger syndrome: Negative
thoughts, behavioral problems, and life interference. Focus on Autism and Other
Developmental Disabilities, 21, 25-35. doi:10.1177/10883576060210010401
Frederickson, N., Simmonds, E., Evans, L., & Soulsby, C. (2007). Assessing the social and
affective outcomes of inclusion. British Journal of Special Education, 34, 105-115.
doi:10.1111/j.1467-8578.2007.00463.x
George, D., & Mallery, P. (2003). SPSS for Windows step by step: A simple guide and reference.
(11.0 update, 4th ed.). Boston, MA: Allyn & Bacon.

163

Gilchrist, A., Green, J., Cox, A., Burton, D., Rutter, M., & Le Couteur, A. (2001). Development
and current functioning in adolescents with Asperger syndrome: A comparative study.
Journal of Child Psychology and Psychiatry, 42, 227-240. doi:10.1007/s10803-0121679-5
Goodenow, C. (1993). Classroom belonging among early adolescent students: Relationships to
motivation and achievement. The Journal of Early Adolescence, 13, 21-43.
doi:10.1177/0272431693013001002
Gresham, F. M., & Elliott, S. N. (1990) Social Skills Rating System Manual. Circle Pines, MN:
American Guidance Service.
Gresham, F. M., & Elliott, S. N. (2008) Social Skills Improvement System Rating Scales Manual.
Minneapolis, MN: Pearson.
Gross, E. F. (2004). Adolescent Internet use: What we expect, what teens report. Applied
Developmental Psychology, 25, 633-649. doi:10.1016/j.appdev.2004.09.005
Harmse, R. G., & Pottas, D. (2010). Help me to understand your world: A reflection on the
potential impact of e-health systems on the prognosis for Asperger Syndrome. E-Health,
335, 210-221. doi:10.1007/978-3-642-15515-4_22
Harper, C. B., Symon, J. B. G., & Frea, W. D. (2008). Recess is time-in: Using peers to improve
social skills of children with autism. Journal of Autism and Developmental Disorders, 38,
815-26. doi:10.1007/s10803-007-0449-2
Hayter, C. A. (2009). Parent and child perspectives on adolescents development with autism
spectrum disorder. (Unpublished thesis). Michigan State University, East Lansing, MI.
Hayter, C. A., Oka, E. R., Tran, T., & Kloss. M. (2012, February). Adolescent voices: Growing
up with autism spectrum disorder. Poster presented at the National Association of School
Psychologists Conference, Philadelphia, PA.
Hedley, D., & Young, R. (2006). Social comparison processes and depressive symptoms in
children and adolescents with Asperger syndrome. Autism, 10, 139-153.
doi:10.1177/1362361306062020
Henry, F. M., Reed, V. A., & McAllister, L. L. (1995). Adolescents’ perceptions of the relative
importance of selected communication skills in their positive peer relationships.
Language, Speech, and Hearing Services in Schools, 26, 263-272. doi:01611461/95/2603-0263
Ingraham, C. L., & Oka, E. R. (2006). Multicultural issues in evidence-based interventions.
Journal of Applied School Psychology, 22, 127-149. doi:10.1300/J370v22n02_07

164

Insel, T. (2013, April 29). Transforming Diagnosis. Retrieved from
www.nimh.nih.gov/about/director/2013/transforming-diagnosis.shtml
Kensinger, E. A. (2011). What we remember (and forget) about positive and negative
experiences. Psychological Science Agenda, 25(10). Retrieved from
http://www.apa.org/science/about/psa/index.aspx.
Knott, F., Lewis, C., & Williams, T. (2007). Sibling interaction of children with autism:
Development over 12 months. Journal of Autism and Developmental Disorders, 37,
1987-1995. doi:10.1007/s10803-006-0347-z
Kochanska, G., Coy, K. C., & Murray, K. T. (2001). The development of self-regulation in the
first four years of life. Child Development, 72, 1091-1111. Retrieved from
http://www.jstor.org/stable/1132431.
Koning, C., & Magill-Evans, J. (2001). Social and language skills in adolescent boys with
Asperger Syndrome. Autism, 5, 23-36. doi:10.1177/1362361301005001003
Kovacs, M. (1992). Children’s Depression Inventory. North Tonawanda, NY: Multi-Health
Systems.
Krasny, L., Williams, B. J., Provencal, S., & Ozonoff, S. (2003). Social skills interventions for
the autism spectrum: Essential ingredients and a model curriculum. Child and Adolescent
Psychiatric Clinics of North America, 12, 107-122. Retrieved from
http://www.elsevier.com/journals/child-and-adolescent-psychiatric-clinics-of-northamerica/1056-4993
Kratochwill, T. R., & Shernoff, E. S. (2004). Evidence-based practice: Promoting evidencebased interventions in school psychology. School Psychology Quarterly, 18, 389-408.
Retrieved from http://www.apa.org/pubs/journals/spq/
Kuusikko, S., Pollock-Wurman, R., Jussila, K., Carter, A. S., Mattila, M.L., Ebeling, H., &
Pauls, D. L. (2008). Social anxiety in high-functioning children and adolescents with
Autism and Asperger syndrome. Journal of Autism and Developmental Disorders, 38,
1697-1709. doi:10.1007/s10803-008-0555-9
La Greca, A. M., & Stone, W. L. (1993). Social Anxiety Scale for Children-Revised: Factor
structure and concurrent validity. Journal of Clinical Child Psychology, 1, 17–27.
doi:10.1207/s15374424jccp2201_2
Lasgaard, M., Nielsen, A., Eriksen, M. E., & Goossens, L. (2010). Loneliness and social support
in adolescent boys with autism spectrum disorders. Journal of Autism and Developmental
Disorders, 40, 218-226. doi:10.1007/s10803-009-0851-z
Laugeson, E. A. & Frankel, F. (2010). Social Skills for Teenagers with Developmental and
Autism Spectrum Disorders: The Peers Treatment Manual. New York, NY: Routledge.

165

Laugeson, E. A., Frankel, F., Gantman, A., Dillon, A. R., & Mogil, C. (2012). Evidence-based
social skills training for adolescents with autism spectrum disorders: The UCLA Peers
Program. Journal of Autism and Developmental Disorders, 42, 1025-1036.
doi:10.1007/s10803-011-1339-1
Laugeson, E. A., Frankel, F., Mogil, C., & Dillon, A. R. (2009). Parent-assisted social skills
training to improve friendships in teens with autism spectrum disorders. Journal of
Autism and Developmental Disorders, 39, 596-606. doi:10.1007/210803-008-0664-5
Lee, H., Marvin, A. R., Watson, T., Piggot, J., Law, J. K., Law, P.A., Constantino, J. N., &
Nelson, S. F. (2010). Accuracy of phenotyping of autistic children based on Internet
implemented parent report. American Journal of Medical Genetics, 153B, 1119-1126.
doi:10.1002/ajmg.b.31103.
Licciardello, C. C., Harchik, A. E., & Luiselli, J. K. (2008). Social skills intervention for children
with autism during interactive play at a public elementary school. Education and
Treatment of Children, 31, 27-37. doi:10.1353/etc.0.0010
Lindner, J. L., & Rosen, L. A. (2006). Decoding of emotion through facial expression, prosody
and verbal content in children and adolescents with Asperger’s syndrome. Journal of
Autism and Developmental Disorders, 36, 769-77. doi:10.1007/s10803-006-0105-2
Lochman, J. E., & Lampron, L. B. (1986). Situational social problem solving skills and selfesteem of aggressive and nonaggressive boys. Journal of Abnormal Psychology, 14, 605–
661. doi:10.1007/BF01260527
Lopata, C., Thomeer, M. L., Volker, M. A., Nida, R. E., & Lee, G. K. (2008). Effectiveness of a
manualized summer social treatment program for high-functioning children with autism
spectrum disorders. Journal of Autism and Developmental Disorders, 38, 890-904.
doi:10.1007/s10803-007-0460-7
Loveland, K. A., Pearson, D. A., Tunali-Kotoski, B., Ortegon, J., & Gibbs, M. C. (2001).
Judgments of social appropriateness by children and adolescents with autism. Journal of
Autism and Developmental Disorders, 31, 367-76. doi:10.1023/A:1010608518060
Macintosh, K., & Dissanayake, C. (2006). Social skills and problem behaviours in school aged
children with high-functioning autism and Asperger’s Disorder. Journal of Autism and
Developmental Disorders, 36, 1065-76. doi:10.1007/s10803-006-0139-5
Magill-Evans, J., Koning, C., Cameron-Sadava, A. & Manyk, K. (1995). The child and
adolescent social perception measure. Journal of Nonverbal Behavior, 19, 151-169.
doi:10.1007/BF02175502
Mazurek, M. O., & Kanne, S. M. (2010). Friendship and internalizing symptoms among children
and adolescents with ASD. Journal of Autism and Developmental Disorders, 40, 15121520. doi:10.1007/s10803-010-1014-y

166

Müller, E., Schuler, A., & Yates, G. B. (2008). Social challenges and supports from the
perspective of individuals with Asperger syndrome and other autism spectrum
disabilities. Autism, 12, 173-90. doi:10.1177/1362361307086664
Murray, D. (2002). Autism and computers. Distance Education: Autism Children. The University
of Birmingham, Birmingham.
Myles, B. S. (2005). Children and youth with Asperger syndrome: Strategies for success in
inclusive settings. Thousand Oaks, CA: Corwin Press.
National Autism Center. (2009). National Standards Report. Retrieved from
http://www.nationalautismcenter.org/affiliates/reports.php
National Professional Development Center. (2011). The NPDC on ASD and the National
Standards Project. Retrieved from http://autismpdc.fpg.unc.edu/content/nationalstandards-project
Neihart, M. (2000). Gifted Children With Asperger’s Syndrome. Gifted Child Quarterly, 44,
222-230. doi:10.1177/001698620004400403
Newman, L. (2007, April). Facts from NLTS2: Secondary school experiences of students with
autism. Menlo Park, CA: SRI International. Retrieved from
www.nlts2.org/fact_sheets/nlts2_fact_sheet_2007_04.pdf
Oren, A., Mioduser, D., & Nachmias, R. (2002). The development of social climate in virtual
learning discussion groups. The International Review of Research in Open and Distance
Learning, 3. Retrieved from http://www.irrodl.org/index.php/irrodl/article/viewArticle/80
Procidano, M., & Heller, K. (1983). Measures of perceived social support from friends and from
family. American Journal of Community Psychology, 11, 1–24. doi:10.1007/BF00898416
Rao, P. A, Beidel, D. C., & Murray, M. J. (2008). Social skills interventions for children with
Asperger’s syndrome or high-functioning autism: A review and recommendations.
Journal of Autism and Developmental Disorders, 38, 353-361. doi:10.1007/s10803-0070402-4
Reichow, B., & Volkmar, F. R. (2010). Social skills interventions for individuals with autism:
Evaluation for evidence-based practices within a best evidence synthesis framework.
Journal of Autism and Developmental Disorders, 40, 149-166. doi:10.1007/s10803-0090842-0
Renty, J. O., & Roeyers, H. (2006). Quality of life in high-functioning adults with autism
spectrum disorder: The predictive value of disability and support characteristics. Autism,
10, 511–524. doi:10.1177/1362361306066604

167

Reynolds, C. R., & Richmond, B. O. (2008). Revised Children’s Manifest Anxiety Scale (2nd ed.).
USA: Western Psychological Services.
Reynolds, W. M. (2002). Reynolds Adolescent Depression Scale (2nd ed.). Odessa, FL:
Psychological Assessment Resources, Inc.
Rubin, E., & Lennon, L. (2004). Challenges in social communication in Asperger syndrome and
high-functioning autism. Topics in Language Disorders, 24, 271. Retrieved from
http://journals.lww.com/topicsinlanguagedisorders/pages/default.aspx
Russell, E., & Sofronoff, K. (2005). Anxiety and social worries in children with Asperger
syndrome. The Australian and New Zealand Journal of Psychiatry, 39, 633-638.
doi:10.1111/j.1440-1614.2005.01637.x
Ryan, R. M. (1982). Control and information in the intrapersonal sphere: An extension of
cognitive evaluation theory. Journal of Personality and Social Psychology, 42, 450-461.
Retrieved from http://www.apa.org/pubs/journals/psp/index.aspx
Sansosti, F. J., & Powell-Smith, K. A. (2008). Using computer-presented social stories and video
models to increase the social communication skills of children with high-functioning
autism spectrum disorders. Journal of Positive Behavior Interventions, 10, 162-178.
doi:10.1177/1098300708316259
Sansosti, F. J., Powell-Smith, K. A., & Cowan, R. J. (2010) High-functioning autism/asperger
syndrome in schools: Assessment and intervention. New York, NY: Guilford.
Schall, C. M., & McDonough, J. T. (2010). Autism spectrum disorders in adolescence and early
adulthood  : Characteristics and issues. Journal of Vocational Rehabilitation, 32, 81-88.
doi:10.3233/JVR-2010-0503
Schmidt, C., Stichter, J. P., Lierheimer, K., McGhee, S., & O’Connor, K. V. (2011). An initial
investigation of the generalization of a school-based social competence intervention for
youth with high-functioning autism. Autism Research and Treatment, 1-11, 589539.
doi:10.1155/2011/589539.
Seidner, L. B., Stipek, D. J., & Feshbach, N. D. (1988). A developmental analysis of elementary
school-aged children’s concepts of pride and embarrassment. Child Development, 59,
367–377. Retrieved from http://www.jstor.org/stable/1130316
Shattuck, P. T., Orsmond, G. I., Wagner, M., & Cooper, B. P. (2011). Participation in social
activities among adolescents with an autism spectrum disorder. PLoS ONE, 6. Retrieved
from www.plosone.org

168

Sherer, M., Pierce, K. L., Paredes, S., Kisacky, K. L., Ingersoll, B., & Schreibman, L. (2001).
Enhancing conversation skills in children with autism via video technology: Which is
better, “self” or “other” as a model? Behavior Modification, 25, 140-158.
doi:10.1177/0145445501251008
Shipman, D. L., Sheldrick, R. C., & Perrin, E. C. (2011). Quality of life in adolescents with
autism spectrum disorders: Reliability and validity of self-reports. Journal of
Developmental and Behavioral Pediatrics, 32, 85-89.
doi:10.1097/DBP.0b013e318203e558
Sigafoos, J., O’Reilly, M., & de la Cruiz, B. (2007). How to use video modeling and video
prompting. Austin, TX: Pro-Ed Inc.
Silver, M., & Oakes, P. (2001). Evaluation of a new computer intervention to teach people with
autism or Asperger Syndrome to recognize and predict emotions in others. Autism, 5,
299-316. doi:10.1177/1362361301005003007
Solomon, M., Goodlin-Jones, B. L., & Anders, T. F. (2004). A social adjustment enhancement
intervention for high functioning autism, Asperger’s Syndrome, and pervasive
developmental disorder NOS. Journal of Autism and Developmental Disorders, 34, 649–
668. doi:10.1007/s10803-004-5286-y
Spence, S.H. (1995). The Social Worries Questionnaire. Windsor: NFER-Nelson.
Spence, S.H. (1998). A measure of anxiety symptoms among children. Behaviour Research and
Therapy, 36, 545–566. Retrieved from http://www.journals.elsevier.com/behaviourresearch-and-therapy/
Stichter, J. P., Herzog, M. J., Visovksy, K., Schmidt, C., Randolph, J., Schultz, T., & Gage, N.
(2010). Social competence intervention for youth with Asperger syndrome and highfunctioning autism: An initial investigation. Journal of Autism and Developmental
Disorders, 40, 1067-1079. doi: 10.1007/s10803-010-0959-1
Stross, R. (2011, July 9). The therapist will see you now, via the Web. The New York Times.
Retrieved from http://www.nytimes.com/2011/07/10/technology/bringing-therapists-topatients-via-the-web.html
Tantam, D. (2000). Psychological disorder in adolescents and adults with Asperger Syndrome.
Autism, 4, 47-62. doi:10.1177/1362361300004001004
Thomas, K. C., Ellis, A. R., McLaurin, C., Daniels, J., & Morrissey, J. P. (2007). Access to care
for autism-related services. Journal of Autism and Developmental Disorders, 37, 19021912. doi:10.1007/s10803-006-0323-7

169

Tolan, P. H., & Dodge, K. A. (2005). Children's mental health as a primary care and concern: A
system for comprehensive support and service. American Psychologist, 60, 601-614. doi:
http://dx.doi.org/10.1037/0003-066X.60.6.601
Tyminski, R. F., & Moore, P. J. (2008). The impact of group psychotherapy on social
development in children with pervasive developmental disorders. International Journal
of Group Psychotherapy, 58, 363-79. doi:10.1521/ijgp.2008.58.3.363
Tynes, B. M. (2007). Internet safety gone wild? Sacrificing the educational and psychosocial
benefits of online social environments. Journal of Adolescent Research, 22, 575-584.
doi:10.1177/0743558407303979
Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the Technology Acceptance
Model: Four longitudinal field studies. Management Science, 46, 186-204. Retrieved
from https://www.informs.org/Pubs/ManSci
Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of
information technology: Toward a unified view. MIS Quarterly, 425-428. Retrieved
from http://www.misq.org/
Vygotsky, L. S. (1978). Mind in society. Cambridge, MA: Harvard University Press.
Wagner, M., Newman, L., Cameto, R., Levine, P., & Marder, C . (2007). Perceptions and
expectations of youth with disabilities. A special topic report of findings from the
National Longitudinal Transition Study-2 (NLTS2). Menlo Park, CA: SRI International.
Retrieved from www.nlts2.org/reports/2007_08/nlts2_report_2007_08_complete.pdf
Walsh, L. (2009). The design of multimedia-based instruction to support the academic and social
development of children with autism. (Unpublished thesis). Waterford Institute of
Technology, Waterford, Ireland.
Wang, S., Parrila, R., & Cui, Y. (2012). Meta-analysis of social skills interventions of singlecase research for individuals with autism spectrum disorders: Results from three-level
HLM. Journal of Autism and Developmental Disorders. doi:10.1007/s10803-012-1726-2
Wentzel, K. R. (1991). Relations between social competence and academic achievement in early
adolescence. Child Development, 62, 1066-1078. Retrieved from
http://www.jstor.org/stable/1131152
Williams White, S., Keonig, K., & Scahill, L. (2007). Social skills development in children with
autism spectrum disorders: A review of the intervention research. Journal of Autism and
Developmental Disorders, 37, 1858-68. doi:10.1007/s10803-006-0320-x
Zimet, G. D., Dahlem, N. W., Zimet, S. G., & Farley, G. K. (1988). The multidimensional scale
of perceived social support. The Journal of Personality Assessment, 52, 30-41.
doi:10.1207/s15327752jpa5201_2

170