Principal Component Analysis can Be Used to Discriminate Between Elite and Sub-Elite Kicking Performance

Abstract

Contemporary descriptions of motor control suggest that variability in movement can be indicative of skilled or unskilled performance. Here we used principal component analysis to study the kicking performance of elite and sub-elite soldiers who were highly familiar with the skill in order to compare the variability in the first and second principal components. The subjects kicked a force plate under a range of loaded conditions, and their movement was recorded using optical motion capture. The first principal component explained >92% of the variability across all kinematic variables when analyzed separately for each condition, and both groups and explained more of the variation in the movement of the elite group. There was more variation in the loading coefficient of the first principal component for the sub-elite group. In contrast, for the second principal component, there was more variation in the loading coefficient for the elite group, and the relative magnitude of the variation was greater than for the first principal component for both groups. These results suggest that the first principal component represented the most fundamental movement pattern, and there was less variation in this mode for the elite group. In addition, more of the variability was explained by the hip than the knee angle entered when both variables were entered into the same PCA, which suggests that the movement is driven by the hip.