Factorial structure and construct validity of an italian version of the Fears of Compassion Scales. A study on non-clinical subjects

Objectives: This study was aimed at investigating the factorial structure and the construct validity of the Italian translation of the Fears of Compassion (FC) Scales in a non-clinical sample (i.e., Fear of Compassion From Others [FCFO], Fear of Compassion Towards Others [FCTO] and Fear of Self-Compassion [FSC]). An exploratory factor analysis was conducted on all the items to investigate the dimensionality of the FC scales. To assess construct validity, correlations between the FC scales and a series of construct-related measures were analysed. Design/Methods: After being translated into Italian using a back-translation procedure, the questionnaire was administered to a community sample of 298 participants (82 males) with a mean age of 24.31 (SD = 8.75), along with self-report measures of psychopathological symptoms, attachment style, self-esteem, satisfaction with life and altruistic behaviour. A behavioural test of altruism was also administered in a subsample of 40 subjects. Results: Thirteen of 38 items did not show adequate psychometric characteristics and thus were removed. The remaining 25 items showed a clear 3-factor solution which explained 48% of the variance. FC Scales were significantly correlated with all the construct-related scales administered in the expected direction, with higher effect sizes for FCFO and FSC than FCTO. Conclusions: Although 13 items were removed, results confirmed the expected three factor solution for the Italian translation of FC scales, and provided new evidence for their construct validity. In this vein, an interesting pattern of correlations emerged with psychiatric symptoms and prosocial behaviour, indicating that FCFO and FSC are more powerful correlates of psychopathology and altruism with respect to FCTO. Practitioner points: Fears of Compassion Scales have been increasingly used in clinical and research settings. The reduced Italian version of the FCS developed in this study is a valid and parsimonious instrument. Fears of receiving compassion from others and from ourselves are more powerful predictors of psychopathological symptoms than fear of giving compassion to others. Fears of Compassion Scales were correlated with both a self-report and a behavioural measure of altruism.