Abstract
Keywords
Social interaction and social skills have historically been a domain of concern for occupational therapy practitioners. These skills were identified initially in 1985 as a performance subsystem of the Model of Human Occupation (Kielhofner, 1985) and are included as a client factor in the Occupational Therapy Practice Framework: Domain and Process (3rd ed.; American Occupational Therapy Association [AOTA], 2014), which defines and guides occupational therapy practice (Griswold & Townsend, 2012). Social skills may be a focus of intervention with populations of any age.
Recent educational legislation requires that behavior and social skills that interfere with educational progress be programmatically addressed and included in the individualized education program (Individuals With Disabilities Education Improvement Act, 2004 [IDEA 2004]; Pub. L. 108–446). Behavior and social skills are also to be addressed in schoolwide and systemwide programs within public education (No Child Left Behind Act of 2001 [NCLB]; Pub. L. 107–110). Occupational therapy practitioners may be members of school system teams that design and implement social skills programs, and practitioners experienced in social skills programming can provide useful information in this endeavor. Four principles that guide current occupational therapy practice are (1) client-centered practice, (2) occupation-centered practice, (3) evidence-based practice, and (4) culturally relevant practice (Schell, Scaffa, Gillen, & Cohn, 2014). The study described in this article sought to provide a client-centered, research-based taxonomy to guide occupational therapy intervention to improve social skills.
Social skills include interpersonal, communication, decision-making, and problem-solving skills and are essential for participation in childhood occupations (Hilton, Crouch, & Israel, 2008; Griswold & Townsend, 2012). Although specific constructs underlying social skills are elusive and past studies have frequently focused on a particular skill or skill set, a general consensus on what social skills encompass was offered by the Council for Exceptional Children (2003):
Social skills are skills that make it possible for children and adolescents to get along with others, to gain acceptance as learning and play partners, and to develop and keep friendship. Social skills include 1) social behaviors, such as maintaining eye contact, taking turns, and asking before taking another child’s play materials; 2) emotional and behavior regulation skills that make it possible for children to inhibit disruptive behavior; 3) social-cognitive processes that children utilize to solve social problems, such as attending to and interpreting social cues to understand others’ intentions; and 4) social knowledge, for example, understanding what it means to be a friend. (p. 1)
Social competence is the ability to use and generalize social skills across settings and is the desired outcome of social skills training. Research has demonstrated that social skills training has positive effects on social–emotional development and behavior in young children with disabilities, and a strong relationship has been shown among social competence, behavior, and achievement in randomized controlled trials (What Works Clearinghouse, 2013). A meta-analysis of the effectiveness of school-based social–emotional behavioral programs found broad benefits, including an increase in social skills and a decrease in antisocial behavior, as well as overall benefits in social skills, antisocial behavior, substance abuse, positive self-image, academic achievement, mental health, and prosocial behavior (Sklad, Diekstra, DeRitter, Ben, & Gravesteijn, 2012). Multiple studies have linked early social skill deficits to psychosocial problems in adulthood, including loneliness, depression, alcoholism, social anxiety, and marital distress (Segrin & Flora, 2000).
Children with disabilities show more deficits in social skills than do their nondisabled peers (Fussell, Macias, & Saylor, 2005; Howlin & Goode, 2000). Specific social patterns are depicted as characteristic of some diagnoses. For instance, social reciprocity deficits are considered a core feature of autism spectrum disorder (ASD; White, Keonig, & Scahill, 2007). Difficulty with social skills has been well documented in diagnostic groups of children with disabilities including autism, learning disabilities, attention deficit disorder (ADD), language impairment, and negative social–emotional behavior (Aro, Eklund, Nurmi, & Poikkeus, 2012; Gutstein & Whitney, 2002; Smith, 2000; Smith & Wallace, 2011).
Social skill deficits are also widely recognized throughout the school population at large. Griswold and Townsend (2012) reported significant differences in the quality of social interaction between children with and without disabilities in the natural context of school, child care, and play situations. Hilton et al. (2008) compared children with high-functioning autism with nondisabled peers and concluded that difficulty with social skills was related to some aspects of out-of-school participation patterns.
Educational legislation has evolved to reflect research on social skills and academic participation (e.g., Americans With Disabilities Act of 1990, Pub. L. 101–336; IDEA Amendments of 1997, Pub. L. 105–117; IDEA 2004; NCLB). NCLB identified behavioral management, social skills development, and transition to adult life as legitimate educational outcomes. The IDEA Amendments of 1997 noted that behavioral issues interfere with student learning and that neither punitive measures nor removal from the academic setting is a constructive solution. Positive intervention strategies and related services that focus on students’ strengths and needs have been recommended (Smith, 2000). NCLB emphasized quality education for all students, including those with disabilities, and recognized the need to address social, emotional, and behavioral needs to enable students to learn. Whole-school initiatives and dynamic models to improve student outcomes have also been supported (Smith, 2000).
IDEA 2004 strengthened school-based mental health initiatives and provided opportunities for timely intervention and preventive approaches. Early Intervening Services and Response to Intervention can be instituted, increased, or adapted when a child is showing signs of difficulty. School failure or diagnosis is no longer the sole criteria for receiving services.
Addressing mental health issues in public education has shifted from an individual deficit model to a public health and preventive model (AOTA, 2014; Koller & Bertel, 2006). Under this model, occupational therapy practitioners can participate with other team members in designing and establishing programs targeted to whole schools, at-risk groups, or individual students, thereby promoting mental health through preventive programming (AOTA, 2009). Intervention and prevention activities may include conducting workshops, adapting the environment, and facilitating supports for students at risk (AOTA, 2008, 2009).
Occupational therapy is a primary public education service for children ages birth to 3 yr and a related service for children ages 3–21 yr. The focus of school-based occupational therapy is to enable children to participate in the learning, play, and social participation inherent in the educational experience (AOTA, 2011). Through participation, children learn skills and competencies, build relationships, and establish social networks, all of which contribute to health and well-being (Law, 2002; Law et al., 2004).
Social skill strengths and weaknesses have been identified through various methods, including peer report, parent report, and teacher observation (Griswold & Townsend, 2012). Although parental participation is recognized as important in acquiring, generalizing, and maintaining social skills, few studies have sought information from parents on what skills should be addressed in social skills programs (Kolb & Hanley-Maxwell, 2003). Also missing from the literature is child-centered identification of social difficulties by the children themselves, corroborated and expanded on by parents or caregivers. A child-centered method of eliciting social skill goals would ground intervention in personal experience and foster children’s awareness of and investment in the process. Including the family in the discussion and decision making adds perspective, clarifies priorities, and ensures that the program, child, and family are working together.
Study Purpose
The purpose of this research was twofold:
To develop a social skills taxonomy to provide a framework for social skill categories and constructs and related behavioral descriptors based on a child-generated and parent-corroborated needs assessment
To apply the social skills taxonomy to school-age children in six diagnostic categories and compare social skill patterns across diagnoses.
Method
Research Design
A mixed-method research design was used for this study. First, we used the qualitative approach of grounded theory to analyze 6,897 child- and family-generated social skill goals. The goal of this approach is to expose concepts and relationships within field-based data for the purpose of generating a theory or model (in this case, a taxonomy) that allows for further empirical testing (Creswell, 1998). Second, using the taxonomy structure, we used the quantitative analysis methods of multivariate analysis of variance (MANOVA) and Tukey post hoc honest significant difference (HSD; Portney & Watkins, 1993) test to explore the social skill patterns of 450 children in six diagnostic categories. The study was deemed exempt from institutional review board approval.
Setting
The study took place in the context of a private nonprofit social skills intervention program in a suburb of a large metropolitan city. The occupational therapy–based program offered up to 25 after-school groups in 4 counties in 2 states. The program operated from 1987 to 2013, and the study was conducted from 2001 to 2006. The children who enrolled were ages 3–18 yr and were from suburban, middle-class families. Payment for the program included self-pay, insurance, or scholarship. Children were referred for social skills intervention because of difficulty with peer interaction identified by school personnel, community professionals, and parents.
All groups were led by experienced, licensed occupational therapists with several years of clinical work experience in pediatrics or mental health and specific experience or training in working with social skills. The program consisted of 8 weekly 1.5-hr sessions in a community location; most children reenrolled in successive programs. The intervention was organized around the collaborative planning, creation, and completion of cooperative projects. Social skills were taught using modeling, coaching, problem solving, and positive reinforcement involving common social scenarios including meeting and greeting, engaging in discussion, eating, playing, and working collaboratively on a project.
Procedure for Question 1
The study used anonymous, archived, social skills goal sheets developed from intake interviews with program enrollees. The therapists who conducted the interviews and guided the development of the social skills goals were experienced social skills group leaders working in the program. They prepared to conduct intake interviews in three ways: (1) by studying the written procedure or watching a training video and observing as the lead therapist (Nancy Kauffman, the first author of this article) conducted interviews; (2) by practicing conducting interviews under the observation of the lead therapist; and (3) by being judged competent in conducting interviews by the lead therapist.
The intake interview followed a standard procedure for all children entering the program. The occupational therapist, child, and parent or caretaker met, and the therapist addressed six scripted interview questions to the child and recorded the child’s responses. The therapist then directed the questions to the family member or caregiver and recorded his or her responses. The therapist encouraged children to express their opinions throughout the process. If a child did not respond to a question, this was recorded. The six questions were as follows:
Tell me about some good things that happen [or go on] at your school.
Now tell me about some not-so-good things at school.
Tell me about your friends.
And now tell me about any kids who are not so friendly.
What activities do you like to participate in?
What do you think your parent [or other caregiver] is wishing we could be helping with or working on with you?
Therapists placed particular emphasis on using positive techniques to elicit spontaneous information and descriptions from children that indicated their awareness of the situation. The questions were designed to elicit information on the following social areas:
Positive and negative aspects of the school community
Current success level in attempts to make friends with peers
Personal goals for the program
Whether the social needs were in acquiring or performing social skills
The family’s view of the enrollee’s social skill needs.
Children were assigned to a group deemed therapeutically suitable for them on the basis of the array of social skills needs and goals of group members and not by a specific skill level. The interviewer framed social concerns into goals that all participants agreed with, and the family signed the goal sheet. These goals provided the focus of intervention, and the archived goal sheets provided the data for the development of the social skills taxonomy.
Development of the Social Skills Taxonomy
The first purpose of the research was to develop a comprehensive taxonomy of social skills using grounded theory methodology, which is designed to explore concepts underlying field-based data for the purpose of theory building (Creswell, 1998). This process required the development of a reliable method for coding the data contained in the goal sheets.
An expert panel comprising four interdisciplinary university faculty with qualitative research expertise was convened. The panel members had professional certification, a doctoral degree, research experience, and more than 20 yr of professional teaching experience. Their professional affiliations were physical therapy, speech–language therapy, occupational therapy (Moya Kinnealey, the second author of this article), and health information management. The physical therapist, an accomplished national expert on qualitative research procedures, led the panel. In addition, two occupational therapy social skills practitioners participated in the expert panel as content experts. One (Kauffman) was a clinician with a master of education degree and more than 15 yr experience in pediatric occupational therapy and social skills intervention. The other had 4 yr experience in pediatric occupational therapy and social skills intervention and was completing a master of science degree in occupational therapy.
The expert panel used open coding category development and axial thematic coding to analyze the data contained in the goal sheets. In open coding, data are analyzed by first breaking the data into small components; then exploring the properties, dimensions, and continuum of the components; and finally grouping them into categories. Axial coding is then used to reassemble the data on the basis of logic and identification of the central phenomena, causal conditions, context, interactions, and consequences (Creswell, 1998). The panel grouped the data into categories and then named and defined the categories, thereby creating a conceptualization or construct.
The expert panel started by analyzing the goal sheets of children diagnosed with ADD. (In this article, we use ADD as a category that includes attention deficit hyperactivity disorder [ADHD].) They collectively open coded the social skill goals of 10 children, identifying all behavioral descriptors and then grouping the descriptors. For example, bragging, overtalking, and no verbal volley were included in the category Verbal Presentation, and meet new people, take social risks, and make friendly overtures were included in the category Interpersonal Relationships. The process continued until saturation—that is, until no new information was being generated from the data and all goal sheets meeting the criteria for inclusion had been coded. The process resulted in identification of seven social skill categories:
Verbal Presentation includes goals addressing the need to modify use of words (content) that interfere with acceptance by others.
Nonverbal Presentation includes goals addressing the need to modify behaviors that interfere with acceptance by others.
Emotional Response includes goals addressing the need to modify emotional reactions to frustrating, new, accidental, or unexpected occurrences or when transiting from one situation to another.
Play includes goals addressing the need to modify behaviors related to playing and working with others.
Awareness of Self and Others includes goals addressing the need to improve conscious consideration and valuing of oneself and other people.
Interpersonal Relationships includes goals addressing the need to initiate and maintain effective relationships with other people.
Feelings About Self includes goals addressing the need to modify level of self-esteem.
Reliability and trustworthiness of the coding process were established by calculating the percentage of agreement among the expert panel members’ categorizations of the goals. Percentage of agreement was established using the coefficient of agreement measure (Portney & Watkins, 1993). Panel members individually scored each goal on a goal sheet. The total number of items on which they were in agreement divided by the number of possible agreements provided the coefficient of agreement.
To establish ongoing coding reliability, expert panel members first independently coded the goals for 10% of the goals sheets for children diagnosed with ADD. Interrater reliability for co-coding was .82 using the coefficient of agreement calculation. Next, a four-member coding team was established, consisting of two expert panel members, an occupational therapist social skills practitioner, and a graduate student. (The team included both authors of this article.) Following training and practice coding, interrater reliability of 80% was achieved for the coding team before individual coding was permitted. This process was repeated in coding all goals for each diagnostic group. Throughout the process, the coding team analyzed and discussed any social skill that was difficult to code within the categories and behavioral indicators to ensure that the model was inclusive. Trustworthiness was ensured by establishing and maintaining coding reliability of 80% and saturation of social skill characteristics. Table 1 outlines the resulting social skills taxonomy, including the seven constructs, their goal descriptions, and behavioral indicators.
Social Skills Taxonomy: Social Skill Category, Goal Description, and Behavioral Indicators
Procedure for Question 2
The second purpose of the study was to apply the taxonomy to children in six diagnostic categories to examine social skill patterns among the diagnostic groups. We examined the goal sheets of 456 enrollees representing six diagnostic categories identified in the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM–IV; American Psychiatric Association [APA], 1994): (1) autism, (2) pervasive developmental disorder–not otherwise specified (PDD–NOS), (3) Asperger syndrome, (4) ADD, (5) learning disabilities–verbal, and (6) nonverbal learning disabilities. The total number of goals for the entire sample was 6,897, with an average of 15 goals per child. Table 2 provides the diagnosis, gender, and age of the children whose goal sheets were included in this part of the study.
Diagnosis, Gender, and Age of Dataset Children (N = 450)
Note. ADD = attention deficit disorder; PDD–NOS = pervasive developmental disorder–not otherwise specified.
Results
We calculated a MANOVA among the diagnostic groups and social skill categories to determine whether significant differences existed among diagnostic groups in the social skill categories represented by selected goals. Results indicated significant F values (p = .001) for all social skill categories, indicating that the means were not equal and that differences existed among the diagnostic groups, as depicted in Table 3.
Means, Standard Deviations, and Multivariate Analysis of Variance Results for Effects of Diagnosis on Social Skill Needs
Note. ADD = attention deficit disorder; LDV = learning disabilities–verbal; M = mean; NVLD = nonverbal learning disabilities; PDD–NOS = pervasive developmental disorder–not otherwise specified; SD = standard deviation.
The Tukey post hoc HSD test is a multiple comparison procedure used to determine which mean comparisons in the MANOVA are significant. The following results were obtained, by social skill category and significance of the mean difference (Mdif) between diagnostic groups:
Verbal Presentation: Children with PDD (M = 3.72) had a higher percentage (p = .000) of goals in this category than children in the other diagnostic groups. Children with Asperger syndrome (M = 2.42, Mdif = 1.170, p = .001) and ADD (M = 2.61, Mdif = 0.998, p = .001) had a significantly higher percentage of goals in this category than children with learning disabilities–verbal.
Awareness of Self and Others: Children with Asperger syndrome (M = 2.59) had a higher percentage of goals in this category than those with PDD (Mdif = 1.438, p = .000) and ADD (Mdif = 727, p = .014), who had a significantly lower percentage of goals in this category.
Emotional Response: Children with Asperger syndrome (M = 2.57) had a significantly higher percentage of goals in this category than those with autism (M = 1.42; Mdif = 1.248, p = .000) and PDD (M = 1.23; Mdif = 1.106, p = .03). Children with PDD had the lowest percentage of goals in this category of any diagnosis.
Play: Children with PDD had a significantly higher percentage of goals in this category than those with Asperger syndrome, ADD, or learning disabilities–verbal (p = .000).
Interpersonal Relationships: Children with learning disabilities–verbal (M = 3.00) and nonverbal learning disabilities (M = 3.00) had a higher percentage of goals in this category than those with other diagnoses. Children with PDD had the lowest percentage of goals in this area (M = 0.39, Mdif = 1.785, p = .000).
Nonverbal Presentation: This goal was reflected in all diagnostic groups, and there were no significant differences among diagnostic groups.
Feelings About Self: Few goals were established in this category for children of any diagnosis.
To determine the relationship between social skill categories and the intervening variable of age, a correlation was calculated between age and social skill categories. Inversely correlated with age were Verbal Presentation (r = −.12, p < .006), Nonverbal Presentation (r = −.11, p < .00), and Play (r = −.20, p < .001). Positively correlated with age were Awareness of Self and Others (r = .17, p < .000), Interpersonal Relationships (r = .28, p < .000), and Feelings About Self (r = .30, p < .000). In contrast, Emotional Response was not correlated with age. Six of the seven social skill categories had significant but not strong correlations with age.
Discussion
The first purpose of the research was to develop a comprehensive model of social skills and behavioral indicators. The resulting child-centered taxonomy of social skill constructs and behavioral indicators contains social skills nomenclature that is widely used, readily recognized, and therefore suitable for school and other community applications. The 6,897 behavioral goals of 456 enrollees within six diagnostic categories informed the taxonomy and are reflected within its parameters. This taxonomy provides a useful framework to facilitate the development of prosocial behavior and social competence programs that support mental health and enable students to benefit from the educational experience and successfully transition to adult life.
Self-Esteem
The taxonomy category Feelings About Self may be a by-product of ineffective social interaction rather than a social skill per se. However, the expert panel identified goals related to how children felt about themselves that could not conceptually be included in other social skill categories. Examples of such goals are to feel less shy when approaching a play situation or to not withdraw when approached by other children. Such goals indicate that self-esteem was a concern related to social skills and needed to be included in the taxonomy. Self-esteem is a self-evaluation of how one feels about oneself and one’s personal worth. It forms throughout childhood based on everyday interactions, experiences, and self-comparisons and continues through adolescence and young adulthood (Erol & Orth, 2011). Children with low self-esteem avoid activities, refrain from offering their opinion, do not initiate interaction, or remain in destructive relationships (Sorenson, 2014). Feelings About Self reflected the fewest number of goals (4%) of all categories, and no significant difference in frequency among diagnostic groups was found. However, this category was moderately positively correlated with age (r = .30, p < .000) and with Interpersonal Relationships (r = .14, p = .001).
Age
Age was correlated with six of the seven social skill categories. A low negative correlation was found with Verbal Presentation, Nonverbal Presentation, and Play, and a moderate positive correlation was found with Awareness of Self, Interpersonal Relationships, and Feelings About Self. These findings suggest that whereas younger children tend to demonstrate social skill difficulty in verbal and nonverbal interaction and play issues, older children tend to demonstrate difficulty in awareness of self and others and interpersonal relationships.
Children’s social skills begin developing in very early childhood through social referencing—that is, reading emotional cues and regulating their own responses to their parents and others in their social situation (Thompson & Lagattuta, 2006). With age and experience with other children, increasingly complex interactions develop. Positive interaction, communication skills, and an increase in vocabulary contribute to the child’s ability to communicate, regulate feelings and behaviors, and develop empathy (McClelland & Tominey, 2009). Friendships become increasingly important from preschool to middle school, and by adolescence social skills are necessary to form and maintain friendships. The inability to form these relationships has been found to result in negative outcomes such as delinquency and psychological problems (Hair, Jager, & Garrett, 2002). Social skills that facilitate adjustment in adolescence include interpersonal skills, conflict resolution, intimacy and prosocial behavior, and the attributes of self-control, behavior regulation, social confidence, and empathy (Hair et al., 2002).
Emotional Regulation
Emotional Response is the social skill category accounting for the largest percentage of goals (18% for the total group). Emotional Response is not correlated with age, suggesting that emotional response is a priority skill at any age. The Council for Exceptional Children’s (2003) description of social skills in children includes emotional and behavioral regulatory skills. Behavioral regulation in response to sensory stimuli, referred to as sensory modulation, has been studied in children with ADHD and Asperger syndrome (Kinnealey, Koenig, & Smith, 2011; Mangeot et al., 2001; Pfeiffer, Kinnealey, Reed, & Herzberg, 2005). Although sensory modulation difficulties are related to emotional and behavioral regulatory difficulties, this area warrants further investigation because other precipitating factors may also contribute to challenges in this area.
Social Skill Needs and Diagnostic Categories
The second purpose of the research was to explore the incidence of social skill goals by diagnostic category. The incidence of social skill need by diagnosis was determined by analyzing the goals of 456 program enrollees in six diagnostic categories. A significantly higher percentage of goals in a given social skill category indicates that a diagnostic group experiences a greater need in that skill category. This incidence provides insight into social skill needs related to diagnosis and may help in prioritizing intervention or prevention efforts. Using this interpretation method, several patterns and insights emerged from the data.
Autism, Asperger Syndrome, and PDD–NOS Diagnostic Changes.
The diagnostic categories used in the study reflect the DSM–IV diagnostic criteria, which were in use from 1994 to 2013, when the DSM–5 (APA, 2013) was published. The diagnosis of PDD in the DSM–IV was based on three areas of impact—(1) social interaction, (2) communication, and (3) behavior—and included PDD–NOS, autism, Asperger syndrome, disintegrative disorder, and Rett syndrome. This study included three of these categories: Asperger syndrome, PDD–NOS, and autism. The DSM–5 provides an updated diagnosis of ASD based on two areas of impact: (1) social communications and interaction and (2) restricted, repetitive patterns of behavior, interest, and activity. These impact areas may occur with or without intellectual impairment and with or without language impairment (Autism Speaks, 2014). Under DSM–5 diagnostic criteria, people formerly diagnosed with Asperger syndrome would be included in the diagnosis of ASD without language impairment. Children diagnosed with PDD–NOS would be considered for the diagnosis of social (pragmatic) communication disorder, which reflects impairment in the ability to change communication to match the context, follow rules, or understand what is not explicitly stated, resulting in functional limitations. The Centers for Disease Control and Prevention (2014) recommended that
individuals with a well-established DSM–IV diagnosis of autistic disorder, Asperger disorder, or pervasive developmental disorder not otherwise specified should be given the diagnosis of autism spectrum disorder. Individuals who have marked deficits in social communication, but who do not otherwise meet the criteria for autism spectrum disorder, should be evaluated for social (pragmatic) communication disorder. (p. 1)
Social Skill Differences Among Diagnostic Categories.
In this study, children with Asperger syndrome demonstrated significantly more deficits related to Awareness of Self and Others and Emotional Response. In contrast, children with PDD–NOS demonstrated more difficulty with Play and Verbal Presentation. Children with autism and Asperger syndrome had social skill needs in initiation and maintenance of Interpersonal Relationships.
Children with learning disabilities, both verbal and nonverbal, had more goals (i.e., greater needs) in the category of Interpersonal Relationships than children with other diagnoses. Kavale and Forness (1996) completed a meta-analysis of the social skills and learning disabilities literature over a 15-yr period. They found that across 152 studies that included 6,353 children with learning disabilities, 75% of the children had social skill deficits. The authors concluded that social skill deficits should continue to be viewed as one of the elements of learning disability. They described the major categories of deficit as skill deficit, performance deficit, and self-control deficit.
In this study, children with ADD did not demonstrate a significantly greater need in any of the social skill categories compared with other diagnostic groups. The largest numbers of goals were in Verbal Presentation, Nonverbal Presentation, and Emotional Response. Madan-Swain and Zentall (1990) reported that social rejection of children with ADHD is correlated more with negative physical and verbal behavior than with high activity level and off-task behavior. In adolescence, ostracism by peers may result from behaviors such as running, talking, interrupting, and disruptive or unsafe behavior, as well as inability to perform tasks that require attention (Cunningham, 2002). A recent study by Sciberras and colleagues (2014) of 179 children with ADHD ages 6–8 yr found that 40% had language problems that affected their social and academic functioning.
Limitations and Future Research
This study has several limitations. First, we used a single source of data: goal sheets developed for a social skills occupational therapy program with enrollees who were predominantly suburban, middle-class children attending community schools with little social or economic diversity. Second, the program participants were not divided equally among diagnostic categories; however, the numbers accurately reflect the program participants over the time studied. Consideration of comorbidities and dual diagnoses was not within the scope of this study.
Finally, the DSM–5 was published in 2013, replacing the DSM–IV. Two conditions used in this study were recategorized: PDD–NOS and Asperger syndrome. PDD–NOS was replaced with ASD, and many children previously diagnosed as PDD–NOS may now meet the DSM–5 criteria for social (pragmatic) communication disorder. In addition, the diagnosis of Asperger syndrome has been replaced with ASD without language impairment. Further research investigating the use of the new taxonomy with the new diagnostic categories is warranted.
Implications for Occupational Therapy Practice
The findings of this study have several implications for occupational therapy practice:
The comprehensive social skills taxonomy described in this article includes seven social skill constructs and behavioral indicators that can be useful for broad-based planning in educational settings.
The strengths of the taxonomy include its basis in child-centered social skill goals generated by the children and their families and its education-friendly nomenclature.
The taxonomy provides a research-based framework for addressing social skills that is derived from grounded theory as well as application to children in varied diagnostic categories based on multivariate analysis. Results are discussed in light of current literature.
This taxonomy can inform broad-based or focused intervention and prevention programs to address social, emotional, behavioral, and mental health issues within school and community settings and to prepare students for life beyond school.
Conclusion
Occupational therapy practitioners in school settings are expected to address behavioral and social–emotional issues that interfere with learning. Social skills training has positive effects on social–emotional development, behavior, and achievement. This article describes a research-based, child-centered, comprehensive taxonomy of social skills to facilitate identification of children with social skill concerns, intervention to address those concerns, and tracking of outcomes in school and community settings.
Footnotes
Acknowledgments
We thank Katherine Shepherd, Barbara Mastriano, and Laurinda Harman for their research expertise in designing and piloting the typology. We also thank Sandra Snopkowski, Michael Yuha, Elizabeth Graalman, Sarah Bliss, and our master of occupational therapy students for their contributions to the application of the typology to specific diagnostic populations. Finally, we thank the wonderful families, children, and therapists of the Collage Social Skills Program.