ABSTRACT
The Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) currently serves
approximately 8 million women, infants, and children monthly. Over half of them receive benefits from the
Supplemental Nutrition Assistance Program (SNAP) as well. WIC-eligible children (WEC) are infants and
children younger than 5 years old with household incomes lower than 185% of the federal poverty level. If their
household income is below 130% of the federal poverty level, they are also eligible for SNAP. WIC and SNAP
dramatically differ in their assistance approach: WIC provides nutritious food packages, while SNAP provides
cash-equivalent benefits to buy foods. However, even with these programs, WIC children still experience
significant nutrition risks: e.g., 29.4% of children aged 2-4 were overweight in 2014. Extensive research
suggests a distortion of food choices among SNAP participants, which could be linked to being overweight.
Therefore, more research is needed to understand the effects of WIC and SNAP participation on WEC’s
nutritional outcomes.
WIC implemented a significant overhaul of the food package in 2009, the first time in four decades.
Several limitations exist in the literature that has evaluated the 2009 WIC revision: First, no national study has
examined WIC participation effects; second, no studies have controlled for selection bias in WIC program
participation, which leads to erroneous estimates of the program’s effects; finally, no studies have examined
the marginal effects of SNAP on WEC’s dietary intake and body weight status. This project’s central aim is to
examine the effects of WIC and/or SNAP participation on WIC-eligible children’s dietary intakes and body
weight status since the 2009 WIC revision. Our main hypothesis is that WIC and SNAP provide complementary
food benefits, and participation in either or both programs promotes healthy diets and weight statuses. We
propose to use the National Health and Nutrition Examination Survey (NHANES) 2009-2014 to examine WIC
and SNAP participation choices in a nationally representative sample of eligible children. To account for
selection bias, we propose three statistical techniques, propensity score (PS), instrumental variable (IV), and
regression discontinuity (RD). PS and IV both have limitations: PS cannot account for unobserved biasing
factors, while IV is often challenged due to difficulty in finding a proper instrumental variable. RD can
theoretically overcome these limitations and take advantage of the design that subjects are assigned into the
treatment or control group based on a cutoff score, which is income eligibility in this project. Although RD is a
well-validated technique and has been applied to program evaluation extensively, it is rarely applied to
WIC/SNAP evaluation. With three complementary methods, we will be able to build the initial step to establish
causality between WIC and/or SNAP participation and nutritional outcomes following the 2009 WIC revision.