have been raised about the use of IRT models in the measurement of food insecurity and, in particular, the use of the Rasch model.
The panel reviewed IRT and related statistical models and discussed their use and applicability to the development of such classifications as food insecurity. The panel recommends modifications of the current IRT methodology used by USDA to increase the amount of information that is used and to make the methodology more appropriate to the types of data that are currently collected using the Food Security Supplement to the Current Population Survey.
The panel reviewed how the latent variable models are estimated and issues of identifiability of these models and how IRT models are used by USDA in the measurement of food insecurity. On the basis of this review, the panel suggests how the models might be used in better ways to accomplish this measurement and recommends a simple way to modify the existing models currently used by USDA to take into account the polytomous nature of the data collected.
USDA bases its annual report and estimates of the prevalence of food insecurity on data collected from the Food Security Supplement to the Current Population Survey. The Household Food Security Survey Module, or a modification of it, is or has been used in several surveys. One of the main objectives of the annual food insecurity measure is to monitor the estimated prevalence of food insecurity, as well as changes in its prevalence over time, at the national and state levels to assess both program policies and the possible need for program development.
After reviewing the key features of selected national surveys—the Current Population Survey, the National Health Interview Survey, the National Health and Nutrition Examination Survey, and the Survey of Income and Program Participation—the panel compared the relative merits of each, for either carrying the Food Security Supplement or conducting research to supplement the information obtained from it. The panel recommends research and testing to understand better the strengths and weaknesses of each survey in relation to the Current Population Survey, leading to the selection of a specific survey vehicle for the Food Security Supplement, or for supplementing that information for research purposes.
Currently, the Food and Nutrition Service in USDA uses trends in the prevalence of food insecurity with hunger based on the HFSSM as a mea-