Abstract Background An increase in overweight and obesity rates among children is a major social problem; however, interpretation and comparability of estimations may be affected by the reference values and cut-off points used. The aim of the study was to assess the prevalence of underweight, overweight, obesity and central obesity in 7-year-old to 10-year-old urban children and to compare the results obtained through various standards of BMI or waist-to-height ratio (WHtR) indicators. Methods The research was conducted on a sample group of 367 girls and 424 boys aged 6.5–10.5 years, randomly chosen from a number of primary schools in Warsaw, Poland. In all participants, basic somatic features were assessed, and based on BMI and WHtR values, participants were then classified according to different standards. The prevalence of underweight, overweight, obesity and central obesity in boys and girls was compared using the chi-square test; fractions obtained from various BMI and WHtR standards were compared through a test for proportions; and the conformity of classification methods was assessed using Cohen’s kappa coefficient. Results Approximately 9 % of girls and 6 % of boys were underweight, 15 % of all participants was classified as overweight, and approximately 4 % of girls and 6 % of boys aged 7–10 were obese. Central obesity was diagnosed in 18.6–20.9 % of all participants, while another 7.1 % of girls and 7.5 % of boys displayed symptoms of excessive fat deposition characterized by elevated body fat percentages. Even though the prevalence estimations varied depending on the standards used, the overall classification compliance reached 86–94 % with a Cohen’s kappa coefficient ranging from 0.676 to 0.841. Conclusion The prevalence of underweight, overweight and obesity among urban children is comparable to estimates for the general population. Of particular concern, in terms of health and proper physical development, is the problem of central obesity that affects one out of five children. The use of reference values representing body fat percentage seems justified as it allows for a more precise diagnosis of weight-related disorders, including the particularly threatening abdominal obesity.