The Self-Compassion Scale (SCS) is currently the only self-report instrument to measure self-compassion. The SCS is widely used despite the limited evidence for the scale’s psychometric properties, with validation studies commonly performed in college students. The current study examined the factor structure, reliability, and construct validity of the SCS in a large representative sample from the community. The study was conducted in 1,736 persons, of whom 1,643 were included in the analyses. Besides the SCS, data was collected on positive and negative indicators of psychological functioning, as well as on rumination and neuroticism. Analyses included confirmatory factor analyses (CFA), exploratory factor analyses (EFA), and correlations. CFA showed that the SCS’s proposed six-factor structure could not be replicated. EFA suggested a two-factor solution, formed by the positively and negatively formulated items respectively. Internal consistency was good for the two identified factors. The negative factor (i.e., sum score of the negatively formulated items) correlated moderately to strongly to negative affect, depressive symptoms, perceived stress, as well as to rumination and neuroticism. Compared to this negative factor, the positive factor (i.e., sum score of the positively formulated items) correlated weaker to these indicators, and relatively more strongly to positive affect. Results from this study do not justify the common use of the SCS total score as an overall indicator of self-compassion, and provide support for the idea, as also assumed by others, that it is important to make a distinction between self-compassion and self-criticism.

Compassion for others and self-compassion are assumed to be closely related concepts. Yet, as they have been mostly studied separately, little is known about their relationship and to what extent they differ or resemble each other with respect to their correlates. This cross-sectional study aimed to gain knowledge on their mean levels, interrelationship, and relationships to psychological well-being and demographic factors. A community sample of 328 adults completed a series of standardized self-report questionnaires to assess compassion for others, self-compassion, depressive symptoms, negative affect, and positive affect. Results showed that compassion for others and self-compassion were not significantly related. Self-compassion was more strongly related to negative and positive indicators of affect than compassion for others. Compassion for others was higher in women than in men, and in low educated individuals compared to higher educated individuals. In contrast, self-compassion was lower in low educated individuals. Future research can build up on these findings to enlarge the understanding of how compassion for others and self-compassion relate and differ from each other.

Self-compassion has shown to be beneficial for individuals’ wellbeing; in particular, it has been associated with lower levels of depressive symptoms. The purpose of this study was to further explore the association between self-compassion, as measured by the Self-Compassion Scale (SCS), and depressive symptoms, in a large representative sample of community adults (n = 734, Mean age = 55.7, SD = 15.2). We examined the association of depressive symptoms with the SCS total score, the SCS six subscales (i.e., self-kindness, common humanity, mindfulness, self-judgment, isolation, and over-identification), and the SCS positive and negative items (referred to as self-compassion and self-coldness, respectively). In addition, we explored the predictive ability of self-compassion, self-coldness, and the SCS six subscales on depressive symptoms both cross-sectionally and over a 1-year period of time. Finally, we sought to test the moderating role of self-compassion on the association between self-coldness and depressive symptoms. Results showed that the SCS negative items and subscales were more strongly related to depressive symptoms than the SCS positive items and subscales. Accordingly, self-coldness was a stronger predictor of depressive symptoms, cross-sectionally and over a 1-year timeframe, when compared with self-compassion. Particularly, the feeling of being isolated was shown to be strongly associated with depressive symptoms. We did not find substantial evidence for a moderating role of self-compassion on the association between self-coldness and depressive symptoms. Future research needs to determine the added value of assessing self-coldness and whether or not it is an essential part of self-compassion.