Rasch Modeling and Differential Item Functioning of the Self-Stigma Scale-Short Version among People with Three Different Psychiatric Disorders

Self-stigma is prevalent in individuals with psychiatric disorders and can profoundly affect people. A unified assessment with sound psychometric properties is needed for evaluating self-stigma across psychiatric conditions. The aim of this study was to examine the psychometric properties of the Self-Stigma Scale-Short version (SSS-S) using Rasch modeling. Six-hundred and twelve participants with substance use disorders (n = 319), attention-deficit/hyperactivity disorder (n = 100), and schizophrenia (n = 193) completed the SSS-S. Rasch results confirmed the unidimensionality of the nine items of the SSS-S. The four-point Likert scale of the SSS-S reflected monotonical increases along the self-stigma continuum. No ceiling or floor effects were detected. Among the three subdomains of the SSS-S, cognitive items appeared to be the most robustly endorsed, and behavioral items were the least endorsed. Two items in the SSS-S displayed differential item functioning across the three diagnoses. Additionally, SSS-S scores showed weak to moderate correlation with depression, anxiety, and stress scale scores. The SSS-S had overall satisfactory psychometric properties. Healthcare professionals may use this assessment to assess self-stigma in multiple psychiatric groups, and information gained may facilitate improved care.

Automated summary

This study examined the psychometric properties of the Self-Stigma Scale-Short version (SSS-S) and found that it has satisfactory reliability and validity in assessing self-stigma in individuals with different psychiatric disorders.