Statistics based on USDA’s nationally representative food security survey, as well as qualitative research, indicate that parents in households with inadequate resources for food generally attempt to shield their children from effects of food insecurity by reducing their own food consumption. There is, however, some uncertainty about whether USDA statistics, which are based on an adult’s proxy report of children’s food security, accurately represent the extent of children’s food insecurity. This study compares self-reported food insecurity of older children and youths (ages 12-17) with self-reported food insecurity of adults in the same households using National Health and Nutrition Examination Survey (NHANES) for 2001-08. The public-use NHANES data are supplemented with confidential household roster data in order to identify children and adults living in the same household.

The findings indicate that children and youths ages 12-17 are shielded to a considerable extent from food insecurity relative to adults in the same household, and that the extent to which they are shielded is greater as the severity of adult food insecurity increases. The odds of youth self-reporting moderately severe food insecurity are about 50 percent below the odds for adults in the same household; the corresponding odds at a more severe level, approximating that of very low food security, are about 80 percent below the odds for adults in the same household. The extent to which youths are shielded relative to adults is less in households headed by cohabiting couples than in those headed by married couples or single adults. Evidence is mixed on how accurately national statistics represent the extent of food insecurity among youth. The most directly comparable measures indicate that the national statistics, based on adult proxy reports, over-represent the extent of food insecurity among youth it by about one-third. Comparisons of other measures suggest the opposite.