Psychometric properties of the Self-Compassion Scale - Short Form in young adults with psychosis: Proceed with...

There is increasing research and applied clinical interest in self-compassion as a protective factor that may foster positive emotional adaptations to psychosis. The Self-Compassion Scale - Short Form (SCS-SF) is a popular and economical measure of self-compassion, but its psychometric properties are not yet validated in populations experiencing psychotic disorders.

The factorial validity, reliability, convergent validity, and criterion validity of the SCS-SF were evaluated in young adults with early psychosis (N = 170). Confirmatory factor analysis was utilized to compare the fit of six a priori structures of the SCS-SF. Additional post-hoc exploratory factor analyses were conducted. Structural equation modeling evaluated the convergent and criterion validity of the best-fitting structure of the SCS-SF with theoretically and clinically relevant outcomes.

All a priori structures exhibited inadequate fit, with some producing serious misspecification. Additional exploratory modeling indicated a two-factor model, which was retained as the preferred factor structure and used to develop an alternate, reduced-item model. The composite reliability of these factors was acceptable. A factor representing uncompassionate self-responding was associated with an array of affective and psychotic symptoms, whereas a compassionate self-responding factor was weakly or inconsistently related to fewer of these.

Findings provisionally support a two-factor structure of the SCS-SF in psychosis, but its factorial validity remains uncertain. For psychosis researchers using the SCS-SF, scoring according to the two-factor approach rather than a total score may be important for deriving an accurate understanding of the effects of self-compassion on outcomes in psychosis.