Acceso usuarios
Roberto Alsasua Santos, Javier Arana Idiakez, Daniel Lapresa Ajamil, María Teresa Anguera Argilaga 
Within the framework of an observational methodology study, we used log-linear analysis to analyze efficiency in elite under-16 basketball from the perspective of three key dimensions: shot position, last pass position, and shot result. The reliability of the observational datasets was confirmed by analysis of interobserver agreement (Cohen’s kappa statistic) and a generalizability theory analysis. The most parsimonious hierarchical model selected using backward elimination in the log-linear analysis was a conditional independence model where the association between last pass position and shot position was independent of shot result and the association between shot position and shot result was independent of last pass position. In this article, we discuss the results corresponding to the model’s first- and second-order effects and show the possibilities offered by log-linear analysis within a systematic observation study to analyze efficiency in basketball.
