A Structural Analysis of the Social Skills Improvement System Rating Scales, Parent Form: Measurement Invariance Across Race and Language Format

Open Access
Schneider, Brian Philip
Graduate Program:
School Psychology
Doctor of Philosophy
Document Type:
Date of Defense:
October 10, 2012
Committee Members:
  • James Clyde Diperna, Dissertation Advisor
  • Robert Leslie Hale, Committee Member
  • Jonna Marie Kulikowich, Committee Member
  • Keith B Wilson, Special Member
  • social skills
  • SSIS
  • cross-cultural social skills assessment
  • measurement invariance
The purpose of the current study was to evaluate aspects of structural validity for the Social Skills Improvement System Rating Scales, Parent Form (SSIS-PF). Data were obtained from the SSIS-PF standardization sample. Confirmatory factor analysis (CFA) was applied to examine the instrument’s first-order and higher-order measurement structures. Resulting baseline measurement models were subsequently analyzed for invariance across two variables. Specifically, measurement invariance was examined as a function of race/ethnicity using subsamples of African American, Latino, and Caucasian children. Invariance then was examined as a function of the language in which rating scales were written (i.e., English or Spanish). For both analyses, multi-sample CFA procedures were used to examine invariance at the configural, metric, and structural levels. Initial analyses provided support for the first-order measurement structure of the SSIS-PF, though there was some evidence of a lack of discriminant validity between select subscales (Cooperation and Responsibility). The instrument’s higher-order measurement structure showed evidence of reduced fit to standardization data relative to the first-order model. Follow-up analysis of the higher-order measurement structure of the SSiS-PF was conducted, and an alternative structure was identified. Results of the invariance analyses with first-order baseline models suggested that the SSIS-PF demonstrates configural, metric, and structural invariance as a function of race/ethnicity. Configural and metric invariance were supported in the language format analysis, but structural invariance was not observed across English and Spanish language groups. Results are discussed in reference to implications for the use of the SSIS-PF as well as broader considerations for cross-cultural social skills assessment.