Ethnic traditional sports participation, cultural identity, and psychological wellbeing among Chinese minority adolescents: a moderated mediation model

Ethnic traditional sports constitute a distinctive vehicle for cultural transmission and physical activity among ethnic minority youth, yet their psychological correlates remain underexplored. This study examined whether cultural identity statistically mediates the relationship between ethnic traditional sports participation and psychological wellbeing among Chinese minority adolescents, and whether this indirect pathway is moderated by perceived social support. A total of 1,086 secondary school students (522 boys, 564 girls, M age = 14.26 years, SD = 1.38) from six schools in Guangxi Zhuang Autonomous Region completed anonymous questionnaires measuring ethnic traditional sports participation, cultural identity, perceived social support, and psychological wellbeing. Moderated mediation analysis was conducted using the PROCESS macro (Model 14) with 5,000 bootstrap resamples. Participation in ethnic traditional sports was positively associated with cultural identity ( β = 0.46, p < 0.001), which in turn was associated with psychological wellbeing ( β = 0.37, p < 0.001). The participation–identity link and the identity–wellbeing link combined to form a significant indirect effect [ β = 0.168, 95% bootstrap CI (0.137, 0.200)]. Perceived social support moderated the second stage of the mediation, such that the association between cultural identity and psychological wellbeing was stronger among adolescents reporting higher social support ( β interaction = 0.08, p < 0.001); the index of moderated mediation was significant [ β = 0.037, 95% bootstrap CI (0.014, 0.060)]. These findings suggest that participation in ethnic traditional sports is associated with minority adolescents’ wellbeing, partly through its association with cultural identity, and that this indirect association is stronger when supportive interpersonal environments are in place. The moderated mediation finding is preliminary rather than robust across all plausible clustering scenarios: class and school identifiers were not retained in the anonymized data set, so clustering could not be modeled directly, and the design-effect sensitivity analysis indicates that the interaction would lose significance at an intraclass correlation of about 0.06 or higher. As the design is cross-sectional, these associations may inform culturally grounded school-based programs for minority youth wellbeing pending verification in longitudinal and experimental designs.