Validierung eines komplexen fußballspezifischen und fußballunspezifischen Tests-Leistungsdiagnostik im Spannungsfeld von Test- und Wettkampfleistung

Abstract

Soccer assessments often lack specificity and complexity. Examinations of validity for soccer assessments, based on the relation between test and match performance, are available although they are scarce. A complex soccer-specific field test which is validated cannot be found in the literature at all. The aim of this follow-up study was to assess the validity of a soccer-specific complexity test (FBKT) and a repeated non-specific sprint test (ST) based on three gold standards: match performance, coach ranking, and league affiliation. A German fourth league soccer team (n = 14) was assessed using FBKT and ST in the pre-season for the second half of the 2010/2011 season. Afterwards, their performance in 17 matches was evaluated using video analysis in order to compare them with the test parameters (e. g., pass/assist ratio, goals, duelling behaviour). Eighteen months later, the league affiliation of all players was identified and the coach evaluated the tested players. With regard to match performance, playing time was proved to be the most suitable parameter for validating the test performance (FBKT, ST). The total and minimum times of ST were the most powerful parameters in relation to playing time and explained 50 % and 46 % of the variance, respectively. Concerning the FBKT, the parameters "sum of all linear sprints" (24 %) and "total time of all activity series with penalty time in round two" (22 %) explained the highest amount of variance. Coach ranking generated significant odds ratios for the minimum time of ST (OR = 6.5; CI: 3.2 - 5.2 × 1015; p = 0.037), and total time of speed dribbling of the FBKT (OR = 1.3; CI: 1.2 - 199.5; p = 0.036) based on a proportional-odds model. With regard to league affiliation, significant odds ratios occurred with the following parameters: minimum time of ST (OR = 15.8; CI: 2.16 - 115.4; p = 0.007); total time of ST (OR = 3.28; CI: 1.31 - 8.21; p = 0.011); and, total time of speed dribbling for the FBKT (OR = 1.3; CI: 1.01 - 1.61; p = 0.044). This study showed that the larger logical validity of soccer-specific complex tests does not necessarily translate to a larger construct validity, as compared to non-specific tests. However, their value is not limited to performance evaluation. The FBKT provides valuable information concerning training process control. It is possible that its low validity is caused by the poor performance of the players in soccer-specific consequences of actions (centres, goals) due to the restricted differentiation of players.