A hierarchical model of factors influencing a battery of agility tests

Aims: To investigate the hierarchical contributions of anthropometry, strength and cognition to a battery of prescriptive and reactive agility tests. Methods: Nineteen participants (mean ± S.D.; age: 22.1 ± 1.9 years; height: 182.9 ± 5.5 cm; body mass: 77 ± 4.9 kg) completed four agility tests: a prescriptive linear sprint, a prescriptive change-of-direction sprint, a reactive change-of-direction sprint, and a reactive linear deceleration test. Anthropometric variables included body fat percentage and thigh girth. Strength was quantified as the peak eccentric hamstring torque at 180, 300, and 60°•s-1. Mean reaction time and accuracy in the Stroop word-colour test was used to assess perceptual and decision making factors. Results: There was little evidence of inter-test correlation with the strongest relationship observed between 10m sprint and T-test performance (r2 = 0.49, P<0.01). Anthropometric measures were not strong predictors of agility, accounting for a maximum 23% (P=0.12) in the prescriptive change-of-direction test. Cognitive measures had a stronger correlation with the reactive (rather than prescriptive) agility tests, with a maximum 33% (P=0.04) of variance accounted for in the reactive change-of-direction test. Eccentric hamstring strength accounted for 62% (P=0.01) of the variance in the prescriptive change-of-direction test. Hierarchical ordering of the agility tests revealed that eccentric hamstring strength was the primary predictor in 3 of the 4 tests, with cognitive accuracy the next most common predictor. Conclusion: There is little evidence of inter-test correlation across a battery of agility tests. Eccentric hamstring strength and decision making accuracy are the most common predictors of agility performance.