Background Whether sport engagement is associated with lower aggression remains contested. Less is known about how sport motivation, that is, reasons for engaging in sport, relates to aggression via emotion- and self-regulatory resources. These psychological pathways are rarely examined within a unified framework. Methods In a cross-sectional survey, 485 students (18–25 years) from a public university completed validated Chinese versions of the Sport Motivation Scale II, the Wong and Law Emotional Intelligence Scale, the Self-Control Scale, and the Brief Aggression Questionnaire. Mediation was tested using Hayes’ PROCESS (Model 6; serial mediation with EI → SC) with percentile bootstrapping (5,000 resamples), controlling for sex and age. Results Sport motivation was negatively associated with aggressive behavior (total effect: B = −0.267, 95% CI −0.342, −0.193; β = −0.303). When emotional intelligence and self-control were included, the direct association remained significant ( B = −0.115, 95% CI −0.188, −0.042; β = −0.131). Indirect effects were significant via emotional intelligence ( B = −0.068, 95% CI −0.109, −0.030), via self-control ( B = −0.038, 95% CI −0.070, −0.008), and through emotional intelligence then self-control (serial indirect: B = −0.047, 95% CI −0.071, −0.028). Indirect pathways accounted for 56.87% of the total association. Sensitivity analyses using alternative operationalizations of sport motivation (SMS-II subscales and autonomous/controlled indices), EI branches, and self-control facets yielded consistent inferences for the serial indirect effect; the direction reversed for raw-scored amotivation. Conclusion In this undergraduate sample, sport motivation was associated with lower aggression partly via a hypothesized serial indirect association through emotional intelligence and self-control (EI → SC), even after adjusting for sex and age and across alternative operationalizations of sport motivation. Given the cross-sectional self-report design, the findings should be interpreted as associational; longitudinal and experimental studies are needed to test temporal ordering and causal mechanisms.
Wang et al. (Thu,) studied this question.