Identification Of Movement Strategies In Vertical Jumps
Refereed Conference Meeting Proceeding
The primary aim of this study is to compare the ability of three commonly used clustering techniques to identify movement strategies within countermovement jumps. A secondary aim is to interpret the identified movement strategies. A hierarchical, k-means using non- and normalized subject scores and an Expectation-Maximization approach using normalized subject scores were examined. The ability to identify movement strategies was measured using the r2-value of a regression model to describe jump height. Clusters of the best clustering solution were examined for differences. Hierarchical clustering utilizing normalized subject scores to generate 4 clusters appears to be the most suitable technique. The generated clusters demonstrated clear defining characteristics.
Symposium of the International Society of Biomechanics in Sports.
The 32nd Conference of the International Society of Biomechanics in Sports
Digital Object Identifer (DOI):
United States of America
Dublin City University (DCU)
Open access repository: