Poster #45 - A Psychometric Analysis of the Social Anxiety Scale for Adolescents Among Youth with Autism

• In Event: 2-055 - Poster Session 06
  In Poster Session: PS 06 Section - Developmental Disabilities

Integrative Statement

Background: Social anxiety is common among adolescents with autism spectrum disorder (ASD) and can exacerbate core ASD symptoms. An ongoing challenge for both research and clinical practice is the assessment of anxious symptomatology in ASD. Despite its widespread use among ASD samples, the Social Anxiety Scale for Adolescents (SAS-A) has not received psychometric evaluation within this population. The present study sought to address this gap in the literature by conducting a psychometric analysis of caregiver and adolescent SAS-A forms in a sample of adolescents with ASD by examining item-level agreement, factor structure, measurement invariance, and discriminant validity.

Method: One-hundred and ninety-seven adolescents with ASD and their caregivers completed the SAS-A as part of participation in a larger RCT of a social skills intervention. Only data collected prior to intervention was analyzed. Adolescents met criteria for ASD on the ADOS-G (total score ≥ 7). Average IQ for this sample was 101.70 (18.65) on the Kauffman Brief Intelligence Test, Second Edition. Adolescents also completed an additional measure of social anxiety (Social Interaction Anxiety Scale; SIAS) and depression (short Mood and Feelings Questionnaire; SMFQ), and caregivers completed a measure of social abilities (Social Responsiveness Scale; SRS), as well as the SMFQ.

Data Analysis: Weighted Cohen’s kappa and polychoric correlations were used to examine caregiver-adolescent agreement for each item. A series of confirmatory factor analyses (CFA) and exploratory factor analyses (EFA) were employed to evaluate the factor structure of caregiver and adolescent reports on the SAS-A. A repeated measures design was used to test a series of measurement invariance models (i.e., configural, metric, threshold, and residual variance) that were compared using chi-squared difference tests and fit comparison for nested models.

Results: Results revealed poor caregiver-adolescent item-level agreement across items (Table 1). A two-factor structure (FNE: fear of negative evaluation; SAD: social avoidance and distress) with three items cross-loading was identified as the final model for both caregiver and adolescent forms (Adolescent CFI: 0.97, TLI: 0.97, RMSEA: 0.07; Caregiver CFI: 0.97, TLI: 0.97, RMSEA: 0.09). While configural invariance was evident between reporters, there was a lack of metric invariance, as indicated by significantly worse model fit due to constrained factor loadings. Finally, there was support for convergent and discriminant validity between the SAS-A and other measure of anxiety (SIAS), depression (SMFQ), and social abilities (SRS) (Table 2).

Discussion: Findings suggest that the SAS-A demonstrates reasonable psychometric properties in an ASD sample. Evidence from the present analyses, however, suggests that the standard three-factor structure (La Greca & Lopez, 1998) does not hold in the present sample of adolescents with ASD. Further, due to a lack of measurement invariance beyond the configural level, factor score and mean-level comparisons between adolescent and caregiver forms is problematic. These findings highlight the need for future work to evaluate the psychometric properties of measures used within an ASD sample in order to make valid inferences and, in turn, to better understand, identify, and evaluate treatment of co-occurring symptoms within ASD.

Authors

• Hillary Katherine Schiltz, Marquette University
  Presenting Author

• Brooke Magnus, Marquette University
  Non-Presenting Author

• Alana McVey, Marquette University
  Non-Presenting Author

• Angela D Haendel, Marquette University
  Non-Presenting Author

• Bridget Dolan, Medical College of Wisconsin
  Non-Presenting Author

• Rachel Stanley, Marquette University
  Non-Presenting Author

• Kirsten Willar, Marquette University
  Non-Presenting Author

• Sheryl Pleiss, Great Lakes Neurobehavioral Center
  Non-Presenting Author

• Audrey Carson, Baylor College of Medicine/Texas Children’s Hospital
  Non-Presenting Author

• Mary Carlson, Marquette University
  Non-Presenting Author

• Christina Murphy, Marquette University
  Non-Presenting Author

• Elisabeth Vogt, Marquette University
  Non-Presenting Author

• Brianna Yund, University of Wisconsin, Milwaukee
  Non-Presenting Author

• Amy Vaughan Van Hecke, Marquette University
  Non-Presenting Author