Abstract
Self-compassion is a trait of having a compassionate attitude toward oneself when experiencing a difficult situation. Recent studies focused on its temporal fluctuation; moreover, Neff et al. (2021) developed a scale to assess the state level of self-compassion. Although its psychometric reliability and validity were confirmed, it was unclear whether it could be adopted in diverse samples. Therefore, we aimed to provide additional evidence using Japanese samples that are recognized as having a high propensity for self-criticism. Study 1 (pre-registered) and Study 2 compared several factor models of the scale and demonstrated that the bifactor model using exploratory structural equation modeling was the best solution. Moreover, the state scale performed better in predicting affect than the trait self-compassion, supporting the incremental validity. Most of the correlations with the trait self-compassion and the positive/negative affect were nearly as expected. In Study 3, an experimental induction of self-compassion indicated that those in the self-compassion condition experienced an improved state self-compassion after the writing tasks, whereas those in the control condition did not. Study 4 replicated these results and revealed that state self-compassion was more easily improved by the writing manipulation than trait self-compassion. Most of the previous research’s results were replicated; furthermore, additional supportive evidence was established across the four studies. We suggest the utility of the state self-compassion scale and discuss some different findings obtained from the Western samples from a cultural perspective.