Social Communication in High School Students on the Autism Spectrum: Examining Profiles, Correlations, and Subgroups

Background: Although all individuals with ASD have social-communication impairments, their social-communication profiles can be quite varied. A deeper understanding of social-communication characteristics and how those relate to other skills and behaviors may be helpful in tailoring intervention approaches for individuals with ASD. This knowledge could also support identifying successful pathways for community involvement and employment in the transition to adulthood. This study will examine relationships between cognitive, academic, social, adaptive, and communication skills in a large sample of high school students on the autism spectrum (n= 545) and explore potential subgroups related to social-communication skills using latent-profile analysis.

Objectives: The objectives of this study are (1) to examine the social-communication profiles of high school students with ASD, (2) to explore subgroups of high school students with ASD in relation to receptive and expressive communication, interpersonal skills, and social abilities, and (3) to examine relationships of social-communication skills with other skills and behaviors (e.g., IQ, adaptive functioning, social participation).

Methods: The participants included a sample of 545 adolescents with autism from 60 high schools in three states from the larger Center on Secondary Education for Students with Autism Spectrum Disorders (CSESA) study. Social-communication profiles were examined using three Vineland subdomains (receptive communication, expressive communication, and interpersonal relations) and the Social Responsiveness Scale (SRS-2). Other characteristics that were examined included: non-verbal IQ (Leiter-3), adaptive functioning (Vineland), and academic performance (Woodcock-Johnson-III). Additionally, the research team collected parent and teacher report of social behaviors and skills.

Descriptive statistics were used to examine the social-communication profiles. The subgroups were explored using latent-profile analysis (LPA), a person-centered analytic method that seeks to subset the members of a sample into smaller more homogeneous profiles. Correlations were used to examine relationships of social-communication characteristics with other characteristics and behaviors.

Results:

The adolescents exhibited a wide range of social-communication characteristics on the Vineland subdomains and SRS-2 (see Table 1). The LPA results indicated three subgroups. Subgroup 1 was low on all Vineland scores and high on the SRS-2; subgroup 2 was in the middle on the Vineland subdomains and the SRS-2; and subgroup 3 was high on the Vineland subdomains and low on the SRS-2. Correlations (see Table 2) revealed that communication characteristics (receptive and expressive) tended to have stronger relationships than the social characteristics (interpersonal relations and SRS-2) with cognitive and academic measures. Further, the social characteristics tended to have stronger correlations with parent and teacher reported social behaviors compared to communication characteristics.

Conclusions:

Adolescents in this sample demonstrated a very wide range of social-communication skills. However, the three groups identified by the LPA appear to represent different ranges of ability on the spectrum of social-communication rather than distinct subgroups. Results indicate that social characteristics of adolescents with ASD are more strongly associated with peer interaction and social participation than communication characteristics; therefore, interventions may need to focus more on social skills. Our results also show a relationship between technology and social-communication skills, which may support technology use during treatment. Clinical implications and alternate conclusions will be discussed.