BACKGROUND: Measurement of heart rate variability (HRV) is a useful method for examining cardiac autonomic control. HRV has been measured in many studies among adults, but research on this topic among children is limited. We therefore defined reference values for a large number of HRV parameters among children. METHODS: The subjects were a population sample of 465 mainly prepubertal children 6-8 years of age. Electrocardiogram (ECG) was recorded at rest, and 1- and 5-min ECG samples were selected for HRV analyses. HRV data were analysed in boys and girls separately to study possible gender differences in HRV parameters. Pearson\'s coefficients for correlation of age, maturity stage, height, weight and body mass index - standard deviation score (BMI-SDS) with HRV parameters were computed to study whether these factors confounded the definition of the reference values for HRV parameters. RESULTS: We found no statistically significant differences in HRV parameters between genders. Therefore, we defined the reference values for all HRV parameters as 5th, 25th, 50th, 75th and 95th percentiles from the ECG samples in all children, but not in boys and girls separately. Age, maturity stage, height, weight and BMI-SDS had weak, if any, associations with HRV parameters, suggesting that they did not confound the definition of the reference values. CONCLUSION: Same reference values for HRV parameters can be used in boys and girls. Although 5-min ECG recordings are preferable for measuring HRV, also 1-min recordings provide reliable data on most HRV parameters, especially those that mainly describe cardiac parasympathetic regulation.