Atopic dermatitis is one of the commonest childhood diseases. Large variations in the prevalence of the disease have been observed in different studies. This variation in the prevalence may be not only from genetic and environmental factors but also from methodology of estimating prevalence of the disease. In this study, Hanifin and UKWP criteria have been used in determining prevalence of the atopic dermatitis in school children. Firstly, the educational areas and schools were chosen by the systematic random sample. Then, the disease prevalence of the first and fifth grade students was determined by considering the parental questionnaires and physical examination. The prevalence of AD was 0.8% with Hanifin and 1.5% with UKWP criteria. The overall prevalence of dermatitis was 7.9%. Therefore, the prevalence of AD in a population varies based on the method used. It is very important for a researcher to use the standard methodology where it is possible or define the disease carefully. There is still some needs in determining more sensitive, specific and practical criteria for the diagnosis of AD.