The predictive role of coach–athlete relationship quality in training engagement and skill development among adolescent basketball players

Objective This study aimed to examine the predictive effect of coach–athlete relationship (CAR) quality on training engagement (TE) and shooting skill improvement (SI) among adolescent basketball players, as well as the potential mediating role of TE in this relationship. Methods A total of 128 basketball players aged 16–18 years (including 83 males) were recruited. The Coach–Athlete Relationship Questionnaire (CART-Q) was used to assess CAR, the Task and Ego Orientation in Sport Questionnaire (TEOSQ) measured TE, and a 100-shot stationary shooting test was used to evaluate SI at both pre- and post-season. Statistical analyses included Pearson correlation, hierarchical regression, and structural equation modeling (SEM). The significance level was set at α = 0.05. Results CAR was significantly positively correlated with both TE (r = 0.52, p < 0.001) and SI (r = 0.38, p < 0.001). After controlling for gender and competition experience, CAR remained a significant predictor of TE ( β = 0.48, p < 0.001) and SI ( β = 0.31, p = 0.002). TE partially mediated the relationship between CAR and SI ( β = 0.14, 95% CI [0.06, 0.24]), accounting for 46% of the total effect. The SEM showed a good model fit (χ 2 /df = 1.86, CFI = 0.95, RMSEA = 0.072). Conclusion A high-quality coach–athlete relationship not only directly enhances training engagement in adolescent basketball players but also indirectly facilitates shooting skill improvement over the course of a season by increasing training engagement.