The Correlation Between Self-compassion and Depression Revisited: a Three-Level Meta-analysis

Objectives This three-level meta-analysis aimed at examining the correlation between self-compassion and depression using culturally diverse samples. We also analyzed self-compassion into its sub-components and checked the cross-cultural consistency of the findings. Methods We reviewed 271 articles that measured self-compassion according to Neff's conceptualization. A three-level meta-analytical framework, which partitioned between-study heterogeneity into within- and between-country levels, was used for all analyses. Meta-regression analyses were used to examine the moderating role of culture, according to the revised Minkov-Hofstede model. Results Our findings indicated that self-compassion was negatively correlated with depression with a large effect size (r= - 0.53). This effect size was closer to the correlation between the negative components of self-compassion and depression, while less close to that of the positive components of self-compassion. The distribution of between-study heterogeneity shared a similar pattern, with positive components of self-compassion showing higher variances across societies. There was evidence showing that the relationships between self-kindness and self-judgment with depression varied across societies varying in flexibility and individualism. Conclusions The effect size and cross-cultural consistency of the correlation between self-compassion and depression vary across different sub-components of self-compassion. Future cross-cultural analyses of self-compassion should pay more attention to this divergence.