The growing number of children suffering from abdominal obesity has resulted in a search for a cheap and reliable method for their detection. The aim of the study was to identify whether the measurements and indicators used to assess this type of obesity are equally suitable for this purpose for each population, and additionally to investigate intergenerational changes in measurements and indicators used to determine the prevalence of abdominal obesity. Three cross-sectional anthropological studies were made in 1983, 2000 and 2010. The analysis included 14,819 children and adolescents aged 3-18 years living in Kraków (Poland). Data on selected characteristics (waist circumference, hip circumference) and indicators [waist-to-hip ratio (WHR), waist-to-height ratio (WHtR), conicity index (CI), body mass abdominal index (BMAI)] were collected and compared between the next series of study. Both waist and hip circumferences have systematically increased. WHR showed no clear upward trend, and the differences have usually been low. WHtR index increased slightly, especially in last decade. In case of CI and BMAI, most children tested in 2010 had the highest values of these indicators. In conclusion, not every indicator of the level of abdominal obesity can be used for each population. WHR or WHtR are not necessarily ideal tools for a population in which secular trends of increasing overall body size occur. In this case it seems that waist circumference is the best measure of abdominal obesity. It is important to examine if a population undergoes intergenerational changes and also what their direction is, so researchers may avoid overestimation or underestimation of the percentage of people affected by abdominal obesity.