Creative Commons source reading · Prediction and forecasting
The association between coaching behaviors and athlete burnout among college athletes: an integrated SEM–ANN model based on self-determination theory
Chen Chen; Hao Wu; Lei Zhang
Originally published in Front Psychol · 2026-06-26
From the article
Selected source text. JournalOS removes markup and shortens excerpts where indicated. The full article provides the complete context.
Abstract
BackgroundAs competitive training pressure continues to rise in higher education, athlete burnout among university students has become more prominent and is often accompanied by lower motivation, reduced performance, and stronger intentions to withdraw from sport. Although prior studies have examined the association between coaching behaviors and athlete burnout, less attention has focused on how this association relates to basic psychological needs, and few studies have combined theory-based testing with predictive analysis. Guided by Self-Determination Theory, this study tested a parallel mediation model including autonomy-supportive and controlling coaching behaviors, autonomy, competence, relatedness, and athlete burnout. The study also examined gender differences.MethodsQuestionnaire data were collected from 800 university…
Discussion
This study found that ASC was negatively correlated with AB. CC was positively correlated with AB. These results support Hypotheses H1 and H5. The findings show that different coaching styles match different burnout patterns. ASC is linked to lower AB. This matches Mossman et al. (2024). When coaches give athletes more choices and listen to their views, athletes often report less emotional exhaustion and less sport devaluation. From the SDT view, these interaction styles often go with higher autonomy satisfaction and stronger behavioral engagement. This pattern fits lower burnout levels (Carroll and Allen, 2021). CC is linked to higher AB. This matches Morales-Sánchez et al. (2020) and Choi (2019). Coaching styles with threats, humiliation, or conditional regard often go with more psychological need frustration and higher burnout. In…
Theoretical implications
First, many earlier studies mainly used linear models. We used a two-stage approach with PLS-SEM and ANN. This is a methodological contribution. PLS is a linear and compensatory model. One predictor can rise when another drops. This may not hold for AB. Positive coaching behaviors may not make up for stress or negative behaviors. ANN is nonlinear and non-compensatory. It can capture complex patterns that linear models miss. ANN can also rank the most important predictors. It reports prediction accuracy. This adds support for theory testing. Second, within the SDT framework, our results show how external training cues, like coaching behaviors, relate to athletes’ experiences of autonomy, competence, and relatedness. These need experiences are systematically linked to emotional exhaustion and burnout. By putting contextual factors and…
Editorial reading guide
- What data created and tested the model, and was evaluation independent?
- What is the cost of a false positive, false negative or out-of-distribution decision?
