A Bayesian network to discover relationships between negative features in sport: a case study of teen players

The purpose of this paper is the study of relationships between negative psychological features in young competitive team players via a Bayesian network (BN). The BN let us describe in graphical terms the dependencies and conditional independencies between variables in a multivariate context, and therefore it let us analyze the relationships between the type of motivational climate and orientation of team players (from two different frameworks: goal orientation theory, and self-determination motivation theory), and we took into account their willingness to accept the use of gamesmanship and cheating at the games, related with somatic and cognitive pre-competitive anxiety. The BN analysis includes three reasoning patterns: evidential reasoning, causal reasoning and intercausal reasoning. We built a BN from a data set composed of twelve psychological variables, which were identified as relevant to this study. The structure and parameters were learnt with TETRAD IV and the BN was implemented in Netica software. Copyright Springer Science+Business Media Dordrecht 2014