Exploratory factor analysis of the Social Responsiveness Scale -- Second Edition in a sample of individuals with autism spectrum disorders
Nelson, Andrew T.
MetadataShow full item record
The Social Responsiveness Scale - Second Edition, School Age Form (SRS-2; Constantino & Gruber, 2012) is a rating scale used for screening for autism spectrum disorders (ASDs), as well as a component of comprehensive diagnostic assessments. While this instrument is widely used in clinical work and in research, there has been little investigation of its psychometric properties completed by those other than the test's authors. The purposes of this study were to examine the factor structure, internal consistency, and screening sensitivity of SRS-2 ratings of 264 students with ASDs between the ages of 4 and 18 years. The ratings were completed by special education teaching staff at a community agency and school for students with developmental disabilities. Prior studies conducted by or with the test's authors found a single-factor solution to the SRS and SRS-2. One prior study conducted independently of the test authors also found a one-factor solution (Wigham, McConachie, Tandos, & Le Couteur, 2012). The present study involved a broad exploratory factor analytic approach using principal axis factoring of the inter-item polychoric correlation matrix. Factor extraction criteria generally converged on a four-factor solution. However, solutions consisting of two to six factors were independently examined for interpretability by four different ASD researchers. All four researchers agreed that the four-factor solution was most interpretable: Factor I - Social Awareness and Competence; Factor II - Restricted and Repetitive Behaviors and Interests; Factor III - Atypical Social Communication; and Factor IV - Social Avoidance. This factor solution generally replicated the solution found by Volker et al. (2014) in a broader developmental disabilities sample. Ordinal alpha internal consistency estimates for scales based on the four factors all met or exceeded standards for screening. A screening classification analysis was conducted using various SRS-2 Total Score cut scores. A cut score of 60 yielded a screening sensitivity of .848, while a cut score of 76 yielded a sensitivity of .398. These results were similar to the results of other studies of the SRS or SRS-2 with other types of raters. Implications for instrument revision and future research are provided.