STATISTICAL CONSIDERATIONS FOR THE USDA FOOD INSECURITY INDEX

This paper reviews the statistical properties of the model used to obtain estimates of the prevalence and severity of poverty-linked food insecurity and hunger in the United States. The U.S. Department of Agriculture has annually sponsored data collection efforts to obtain information on food insecurity and hunger since 1995. The assessment of household food insecurity is based on a one-parameter logistic item response model, also referred to as a Rasch model, and applied to a series of 18 questions reported in the Current Population Survey Food Security Module. The paper was used as the basis for discussions concerning future directions of research on the food insecurity measure. This report was originally released in July 1999