Construct validity of percentage of predicted adult height and BAUS skeletal age to assess biological maturity in academy soccer

Abstract

Background The assessment of biological maturity status plays an important role in talent identification and development programs. Aim To compare age at predicted adult height and BAUS skeletal age as indicators of biological maturity status in youth soccer players using a construct-validity approach. Subjects and methods Participants were 114 players from the U12 to U17 age groups of a professional youth soccer academy. Maturity status was determined via percentage of predicted adult height based upon the Khamis-Roche method (somatic maturity) and assessed via the SonicBone BAUSTM system (skeletal maturity). Convergent and known-groups validity were evaluated between maturity assessment methods and by comparing maturity-related selection biases across age groups. Results Although maturity status indicators were largely interrelated (r = .94, 95%CL 0.91–0.96), concordance (κ = 0.31 to 0.39) and Spearman’s rank-order correlations (ρ = 0.45–0.52) of classification methods were moderate. A selection bias towards early maturing players emerged in the U14 age group which remained relatively consistent through to the U17 age group. Conclusions Results confirm the construct-validity of both methods to assess biological maturity status although further validation relative to established indicators of biological maturity is needed. Furthermore, caution is also warranted when interpreting maturity status classification methods interchangeably given the poor concordance between classification methods.