ther research concerning factors that influence the decision to apply for and participate in WIC should be conducted.

The panel has not had time to fully consider alternative methods but does propose a preliminary alternative. The method applies WIC participation rates from the latest year available to estimates of eligibility for the upcoming year. For example, to estimate participation for 2002, the number of people who participated in WIC from the most recent year with available data, divided by the number of people estimated to be eligible for that year, would be used to adjust the estimates of eligible people for 2002. Since it is likely that participation for each of the eligibility categories varies (e.g., children’s participation rates may be lower than infants’ participation rates), separate adjustments for each eligibility category should be made. This measure is conceptually easy to grasp and can be constructed with existing data.² The merits and drawbacks of this method need to be further explored, and its predictive value should be assessed. Further work could also explore the use of a more sophisticated method that attempts to control for the business cycle or for population composition between the lagged year and the prediction year. But in the short run, the lagged WIC participation rate has promise as an alternative to current practice.

In the long run, the Food and Nutrition Service should sponsor more research on WIC participation decisions and behavior. Program participation modeling studies, such as those that have been conducted for other social welfare programs (Blank and Ruggles, 1994; Currie and Gruber 1996a, 1996b; Moffitt, 1992), could also be applied to the WIC program. Descriptive studies could also be valuable in building a base of knowledge about WIC participation. Studies that explore trends in participation in WIC such as those that are conducted for the Food Stamp Program (see Castner, 2000; and Castner and Cody, 1999, for recent publications on a series of food stamp participation reports) are one example.