Self-compassion and cultural values: A cross-cultural study of self-compassion using a multitrait-multimethod (MTMM) analytical procedure

Self-compassion is natural, trainable and multi-faceted human capacity. To date there has been little research into the role of culture in influencing the conceptual structure of the underlying construct, the relative importance of different facets of self-compassion, nor its relationships to cultural values. This study employed a cross-cultural design, with 4,124 participants from 11 purposively sampled datasets drawn from different countries. We aimed to assess the relevance of positive and negative items when building the self-compassion construct, the convergence among the self-compassion components, and the possible influence of cultural values. Each dataset comprised undergraduate students who completed the "Self-Compassion Scale" (SCS). We used a confirmatory factor analysis (CFA) approach to the multitrait-multimethod (MTMM) model, separating the variability into self-compassion components (self-kindness, common humanity, mindfulness), method (positive and negative valence), and error (uniqueness). The normative scores of the Values Survey Module (VSM) in each country, according to the cultural dimensions of individualism, masculinity, power distance, long-term orientation, uncertainty avoidance, and indulgence, were considered. We used Spearman coefficients (r_s) to assess the degree of association between the cultural values and the variance coming from the positive and negative items to explain self-compassion traits, as well as the variance shared among the self-compassion traits, after removing the method effects produced by the item valence. The CFA applied to the MTMM model provided acceptable fit in all the samples. Positive items made a greater contribution to capturing the traits comprising self-compassion when the long-term orientation cultural value was higher (r_s = 0.62; p = 0.042). Negative items did not make significant contributions to building the construct when the individualism cultural value was higher, but moderate effects were found (r_s = 0.40; p = 0.228). The level of common variance among the self-compassion trait factors was inversely related to the indulgence cultural value (r_s = -0.65; p = 0.030). The extent to which the positive and negative items contribute to explain self-compassion, and that different self-compassion facets might be regarded as reflecting a broader construct, might differ across cultural backgrounds.