Abstract
Introduction. Rugby is one of the most physically and mentally taxing collision sports, not only for players but also for coaches. Over the years, the development of athletes has increasingly incorporated psychological skills training. However, very few instruments exist to assess the psychological skills of rugby coaches, particularly in Malaysia. This study aimed to develop and evaluate the Psychological Skills Training (PST) Index for rugby coaches by defining competency rubrics using pattern recognition techniques and classifying coaches’ readiness to apply these rubrics.
Methods. A cross-sectional survey was conducted among 109 rugby coaches working at school, district, state, and national levels across the country. A 30-item PST instrument was administered, covering the constructs of perceived usefulness, satisfaction, and ease of implementation. The following psychometric procedures were applied: Cronbach’s alpha, the Kaiser–Meyer–Olkin (KMO) measure, exploratory factor analysis (EFA), and discriminant analysis (DA). Additional machine-learning classifiers were used to evaluate the PST Index and determine its predictive performance.
Results. The PST Index demonstrated strong internal reliability and sampling adequacy (α = 0.878; KMO = 0.722). Factor analysis extracted six factors from 13 items, with loadings ranging from 0.600 to 0.770, explaining 46.4% of the total variance. Coaches’ group mastery accuracy indicated that the model could discriminate among low, moderate, and high groups with 96.3% accuracy. Predictive validity was supported by logistic regression (AUC = 0.875) and was similarly supported by neural networks (AUC = 0.874).
Conclusion. The PST Index is the first and currently the only index designed for rugby coaches in Malaysia, and it represents the first index of its kind in Asia. It includes both diagnostic and prognostic capabilities related to psychological readiness and provides predictive utility through pattern recognition. Potential links to injury prevention or coaching stress regulation remain theoretical and were not tested in this study. The PST Index may help educational and coaching institutions substantially refine coaching practices. Further cross-cultural research is needed to validate digital and longitudinal approaches that could increase accessibility and scalability. In this study, validation was limited to internal consistency and internal structure; external validity, longitudinal outcomes, and practical effects were not established.
Key words
psychological skills training, rugby coaches, pattern recognition, psychometric validation, machine learning classification