Construct validity of tests that measure kick performance for young soccer players based on cluster analysis: exploring the relationship between coaches rating and actual measures

Abstract

BACKGROUND: The main aim of this study was to verify the relationship between the classification of coaches and actual performance in field tests that measure the kicking performance in young soccer players, using the K-means clustering technique. METHODS: Twenty-three U-14 players performed 8 tests to measure their kicking performance. Four experienced coaches provided a rating for each player as follows: 1: poor. 2: below average; 3: average; 4: very good: 5: excellent as related to three parameters (i.e. accuracy, power and ability to put spin on the ball). RESULTS: The scores interval established from k-means cluster metric was useful to originating five groups of performance level, since ANOVA revealed significant differences between clusters generated (P<0.01). Accuracy seems to be moderately predicted by the penalty kick, free kick, kicking the ball rolling and Wall Volley Test (0.44 <= r <= 0.56), while the ability to put spin on the ball can be measured by the free kick and the corner kick tests (0.52 <= r <= 0.61). Body measurements, age and PHV did not systematically influence the performance. The Wall Volley Test seems to be a good predictor of other tests. CONCLUSIONS: Five tests showed reasonable construct validity and can be used to predict the accuracy (penalty kick, free kick, kicking a rolling ball and Wall Volley Test) and ability to put spin on the ball (free kick and corner kick tests) when kicking in soccer. In contrast, the goal kick, kicking the ball when airborne and the vertical kick tests exhibited low power of discrimination and using them should be viewed with caution.