Perchlorate (ClO(4)(-)) is ubiquitous in the environment and inhibits the thyroid's uptake of iodide. Food and tap water are likely sources of environmental exposure to perchlorate. The aim of this study was to identify significant dietary sources of perchlorate using perchlorate measured in urine as an exposure indicator. Sample-weighted, age-stratified linear regression models of National Health and Nutrition Examination Survey (NHANES) 2001-2008 data (n=16,955 participants) characterized the association between urinary perchlorate and the mass consumed in USDA food groups, controlling for urinary creatinine and other potential confounders. Separate models of NHANES 2005-2006 data (n=2841) evaluated the association between urinary perchlorate and perchlorate consumed via residential tap water. Consumption of milk products was associated with statistically significant contributions to urinary perchlorate across all age strata: 2.93 ng ClO(4)(-)/ml per kg consumed for children (6-11 years-old (YO)); 1.54 ng ClO(4)(-)/ml per kg for adolescents (12-19 YO); and 0.69 ng ClO(4)(-)/ml per kg for adults (20-84 YO). Vegetables were a significant contributor for adolescents and adults, whereas fruits and eggs contributed significantly only for adults. Dark-green leafy vegetables contributed the most among all age strata: 30.83 ng ClO(4)(-)/ml per kg for adults. Fats, oils, and salad dressings were significant contributors only for children. Three food groups were negatively associated with urinary perchlorate: grain products for children; sugars, sweets, and beverages for adolescents; and home tap water for adults. In a separate model, however, perchlorate consumed via home tap water contributed significantly to adult urinary perchlorate: 2.11E-4 ng ClO(4)(-)/ml per ng perchlorate in tap water consumed. In a nationally representative sample of the United States 6-84 YO, diet and tap water contributed significantly to urinary perchlorate, with diet contributing substantially more than tap water.