Resumo

SASTRE, V et al. Combat analysis in initiation to karate: decision trees in observational methodology. CPD [online]. 2022, vol.22, n.2, pp.247-257.  Epub 12-Maio-2023. ISSN 1989-5879.

This work analyze the technical-tactical performance in combat in the initiation to karate and show the possibilities of application of the decision tree analysis technique in observational methodology. The observational sampling is made up of 887 offensive technical-tactical actions executed in 40 combats by participants of the 8-9 age group. Regarding the recording and coding process, the observation instrument has been introduced in the software LINCE. The quality of the data has been guaranteed by means of an inter-observer concordance procedure based on Cohen's Kappa coefficient and by a design of generalizability. Two decision trees have been made using two different algorithms (CHAID and CART). The results obtained with both algorithms coincide in pointing out the "segment" dimension as the best predictor of the "body area to which the technique is directed"; but they differ in the second best forecaster: dimension "technique group" through the CHAID algorithm, and "concrete technique" dimension with the CART algorithm.

Palavras-chave : kumite; combat analysis; hierarchical segmentation technique; classification tree.

        · resumo em Português | Espanhol     · texto em Espanhol     · Espanhol ( pdf )