Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
In this paper we analyze participation in the National School Lunch Program (NSLP) using a basic neoclassical demand model. This statistical analysis incorporates nutrient consumption from nonlunch sources to control explicitly for tastes. Because the dependent variable takes only four values, the ordered probit estimation technique is used. The insights gained concerning the effects of tastes, the size of the Food Stamp program bonus, and the price of the school lunch, among others, should prove useful in designing future NSLP policies.