There is an increasing need for suitable tools to evaluate body composition in paediatrics. The Body Composition Monitor (BCM) shows promise as a method, but reference values in children are lacking. Twenty children were included and measured twice by 4 different raters to asses inter- and intra-rater reproducibility of the BCM. Reliability was assessed using the Bland-Altman method and by calculating intraclass correlation coefficients (ICCs). The intra-rater ICCs were high (≥ 0.97) for all parameters measured by BCM as were the inter-rater ICCs for all parameters (≥ 0.98) except for overhydration (0.76). Consequently, a study was set up in which BCM measurements were performed in 2058 healthy children aged 3–18.5 years. The age- and gender-specific percentile values and reference curves for body composition (BMI, waist circumference, fat mass and lean tissue mass) and fluid status (extracellular and intracellular water and total body water) relative to age were produced using the GAMLSS method for growth curves.Conclusion: A high reproducibility of BCM measurements was found for fat mass, lean tissue mass, extracellular water and total body water. Reference values for these BCM parameters were calculated in over 2000 children and adolescents aged 3 to 18 years. What is Known• The 4-compartment model is regarded as the ‘gold standard’ of body composition methods, but is inappropriate for regular follow-up or screening of large groups, because of associated limitations. • Body Composition Monitor® is an inexpensive field method that has the potential to be an adequate monitoring tool.What is New• Good reproducibility of BCM measurements in children provides evidence to use the device in longitudinal follow-up, multicentre and comparative studies.• Paediatric reference values relative to age and sex for the various compartments of the body are provided.