This study examined the reliability and validity of the 20-meter shuttle test as a predictor of peak VO2 in Edinburgh school children. Thirty-three children (15 boys, 18 girls) performed three shuttle tests and three laboratory treadmill tests of peak VO2. Multiple regression analysis revealed that the prediction of peak VO2 (ml·kg−1·min−1) from shuttle run performance was improved by including skinfold thickness measurements in the prediction models, particularly with the female group. Predictive power was greatest for females when using maximal shuttle speed (kmhr−1) best of three repeat tests and triceps skinfold thickness (mm) (R2 = .85, SEE = 2.4), and for males when using maximal shuttle speed (km·hr−1) best of three repeat tests and the sum of the triceps and subscapular skinfolds (R2 = .68, SEE = 3.23). When using shuttle run performance to predict peak VO2 (ml·kg−1·min−1) in children of this age group, body composition measures must be included in the equation.