(PS10-45) Beyond the Self in Self-compassion: Interdependent Values as a Moderator in the Mediated Relation Between Social Support, Self-compassion, and Mental Health

Abstract Text: Self-compassion is linked to better mental health outcomes in samples of POC (e.g., Blanden et al., 2021; Chahar Mahali et al., 2021; Zhang et al., 2019). However, self-compassion in Western frameworks is often viewed through an individualistic lens that may not be culturally responsive to the needs of POC. Few studies have explored how relational factors influence the development of self-compassion. Research suggests receiving support from others may promote one’s ability to be kind to oneself in the face of suffering (Toplu-Demirtaş et al., 2018). Thus, the present study examined whether self-compassion mediates the relation between social support and mental health. Given that values of collectivism and interconnectedness are frequently shared among communities of color (e.g., French et al., 2020; Kitayama et al., 2022) and may impact how one relates to oneself and others, we investigated whether interdependent values moderate the association between social support and self-compassion.
Participants included 238 adults who self-identified as “people of color” recruited from an urban and racially diverse university. Self-compassion was measured by the Self-Compassion Scale Short Form (Raes et al., 2011). Social support (latent factor) consists of items from the 2-Way Social Support Survey - Receiving Emotional Support subscale (Shakespeare-Finch & Obst, 2011) and the Perceived Social Support Questionnaire - Brief Form (Lin et al., 2019). Interdependent values (latent factor) consists of items from the Relational-Interdependent Self-Construal Scale (Cross et al., 2000) and the Individualism-Collectivism Interpersonal Assessment Inventory (Matsumoto et al., 1997). Mental health symptoms (latent factor) consist of subscales from the Depression, Anxiety, and Stress Scale-21 (Lovibond & Lovibond, 1995). Well-being (latent factor) consists of subscales from the Mental Health Continuum – Short Form (Lamers et al., 2011).
CFA results suggested a good fit to the data with mental health symptoms as the outcome: χ² (30) = 70.65, p= <.00, CFI = .978, TLI = .967, RMSEA = .078 (90% CI = 0.054– 0.101), SRMR = .041. All observed variables significantly loaded onto their respective factors (λs = .791 - .961). CFA results suggested an excellent fit to the data with well-being as the outcome: χ² (30) = 44.58, p=.042, CFI = .992, TLI = .988, RMSEA = .046 (90% CI = 0.009– 0.073), and SRMR = .031. All observed variables significantly loaded onto their respective factors (λs = .727 - .960). We conducted latent variable regression to examine the indirect effect of social support on mental health outcomes through self-compassion. Results revealed that self-compassion partially mediated the association between social support and mental health symptoms (indirect effect = -0.140, SE = 0.034, p< 0.001; proportion mediated = 0.381, SE = 0.089, p< 0.001), and the association between social support and well-being (indirect effect = 0.103, SE = 0.024, p< 0.001; proportion mediated = 0.222, SE = 0.046, p< 0.001). The moderating effect of interdependent values on the association between social support and self-compassion was not statistically significant. Results will be discussed in terms of implications for intervention and directions for future research.