Measurement invariance of the factors.

This study examined adolescent basketball participation through the lens of Self-Determination Theory (SDT), focusing on the roles of passion (harmonious and obsessive) and affect (positive and negative), with gender and age as moderators. We collected data from 400 Chinese adolescents from south of China using convenience sampling (M age = 15.20, SD = 1.99). Participants completed the Passion Scale and Positive and Negative Affect Schedule Scale (PANAS). Basketball participation was operationalized as self-reported engagement in basketball over the past seven days. Hypothesized pathways were tested using structural equation modeling (SEM) in Mplus, which included test of measurement invariance and moderation. Results revealed that harmonious passion positively predicted basketball engagement (β = .280, 95% CI [.127,.432], p = .003) though enhanced positive affect. Specifically, harmonious passion was positively associated with positive affect (β = .495, 95% CI [.356,.635], p p = .001). In contrast, obsessive passion was linked to negative affect (β = .196, 95% CI [.083,.310], p = .004) but showed no direct effect on basketball participation. While no significant gender differences were observed, age differences emerged. Specifically, the pathways from both harmonious and obsessive passion to basketball engagement were stronger among junior high school students compared to their senior high school counterparts. This study revealed that harmonious passion and positive affect play pivotal roles in sustaining adolescent basketball participation, whereas obsessive passion’s negative emotional consequences may undermine engagement. The findings underscore the importance of interventions designed to promote harmonious passion and positive affect, particularly for senior high students facing academic pressures. Given the cross-sectional design and sampling from southern China, the generalizability of our findings to other cultural contexts or regions of China should be interpreted with caution. Additionally, due to the less-than-ideal model fit in multi-group analyses examining gender as a moderator, results concerning gender differences require the same cautious approach.