Background In order to obtain background references when dealing with serum eosinophil cationic protein (s-ECP) measurements in children with allergic diseases, population-based studies are important. The objectives of our study were to explore the strength of associations between the s-ECP level and atopic dermatitis (AD), allergic rhinitis (AR) and asthma in an unselected northern Norwegian schoolchildren population. Methods s-ECP was sampled from 396 schoolchildren aged 7-12 years from Sor-Varanger community, northern Norway as a part of a population-based study of allergy. In advance, anamnestic information concerning a history of AD, AR and asthma were obtained. The children underwent a clinical investigation, including skin prick tests and peak expiratory flow measurements, where the presence of AD, AR and asthma were evaluated. The associations of these diseases to the s-ECP values were examined in bivariate statistical analysis. Results No statistical significant associations were detected in bivariate analysis between s-ECP and AD, AR or asthma: the mean s-ECP in children without self-reported AD/AR/asthma was 4.6 mug/L [95% confidence interval (CL) 4.0-5.2]. The mean s-ECP in children with self-reported AD or AR or asthma was 5.2 mug/L (95% CI 4.1-6.2), 4.6 mug/L (95% CI 3.5-5.7) and 6.4 mug/L (95% CI 4.4-8.3), respectively. The highest mean s-ECP level was measured in children with clinically diagnosed asthma; 7.1 mug/L (95% CI 4.0-10.3). Above the 75-percentile level of s-ECP, only 17.2% of the children had a history of asthma. Conclusions In this unselected children population, the occurrence of AD or AR was not reflected by an increase in the s-ECP level. The s-ECP was increased in children with asthma, but was not statistically significant. Furthermore, the majority of children with high s-ECP values were not asthmatics. We conclude that the associations between s-ECP and allergic diseases are weak in an unselected population of children.