Assessment of Athletic Development in Youth Players – Goal Setting with Normative Data from Basketball

INTRODUCTION The nationwide implementation of physical performance test batteries for youth squad players can be valuable for compiling individual physical performance profiles based on age- and gender-specific norm values. This approach is frequently used for optimizing training prescription and thus athletic development. The aim of this study was to introduce a distribution-based approach to derive an effect size scale for assessing athletic development from normative testing data in youth players, which can then be translated to setting performance goals for athletic development.METHODS Secondary analysis of norm values (mixed longitudinal and cross-sectional data [1]). In the age-groups under 12 to under 17, a maximum number of 1,172 and 846 tests were available for male and female basketball squad players, respectively. Biannual testing was conducted as part of a federal research project (20-m sprint, 20-m change of direction sprints with/without basketball, jump & reach, standing long jump, chest pass, mid-range jump shot, multistage fitness test). An effect size scale was derived from norm values which were available as quintile scores (five categories). Trivial changes were defined as the age-related mean annual performance development which was estimated as the average age-group-to-age-group change for the quintiles. Threshold values for small, medium, and large changes were calculated as average changes that were required to increase performance classification by one, two or three categories, respectively. These thresholds were additionally compared to the default effect size scale commonly used for interpreting standardized mean differences (between-player standard deviation: small 0.2, medium 0.6, large 1.2 [2]).RESULTS For example, the age-related mean annual development in the jump & reach for male players was 4 cm (trivial change). To reach one, two or three higher performance categories, jump height must improve by 8, 12 and 15 cm, respectively (i.e., small, medium, large). Compared with the default standardized effect size scale, these quintile-based thresholds were larger.CONCLUSION The quintile-based analysis presents a simple and practical approach to derive effect size thresholds based on norm values created from regular physical performance testing. These effect size scales can be easily visualized and communicated to players and coaches, as they are typically familiar with percentile-based performance classification of testing data. A limitation of this study was that only norm values in the form of quintile scores were used for analysis. Future research should attempt to model longitudinal datasets while accountingfor within- and between-player effects. Furthermore, the choice of appropriate and realistic percentilebased thresholds clearly remains up for debate and requires adequate analysis of original longitudinal data.REFERENCES 1. Stadtmann (2012) PhD thesis, Ruhr University Bochum. 2. Hopkins et al. (2009) MSSE,41,3-12.