Although children are exposed to a variety of environmental hazards, including pesticides, there is a scarcity of information available to estimate exposures realistically. This article reports on one of the first attempts to measure multi -pathway pesticide exposures in a population -based sample of urban and nonurban children. A design strategy was developed to assess multi -pathway pesticide exposures in children using personal exposure measurements in combination with complimentary measurements of biological markers of exposure, concentrations in relevant environmental media, and time spent in important microenvironments and participating in exposure -related activities. Sample collection and analysis emphasized measurement of three insecticides ( i.e., chlorpyrifos, diazinon, and malathion ) and one herbicide ( i.e., atrazine ) . These compounds were selected because of their frequent use, presence in multiple environmental media, expected population exposures, and related hazard / toxicity. The study was conducted during the summer of 1997 in Minnesota and involved a stratified sample of households with children ages 3 ± 12 years. Participants resided in either ( a ) the cities of Minneapolis and St. Paul ( urban households ) , or ( b ) Rice and Goodhue Counties just south of the metropolitan area ( non -urban households ) . Results from a residential inventory documenting storage and use of products containing the target pesticides were used to preferentially select households where children were likely to have higher exposures. The study successfully obtained pesticide exposure data for 102 children, including measurements of personal exposures ( air, hand rinse, duplicate diet ) , environmental concentrations ( residential indoor / outdoor air, drinking water, residential surfaces, soil ) , activity patterns ( obtained by questionnaire, diary, videotaping ) , and internal dose ( metabolites in urine ) . Journal of Exposure Analysis and Environmental Epidemiology ( 2000 ) 10, 145 ± 158.