2019 Research Grants

Breakfast in the classroom, body mass index, and academic outcomes.

PI: Michael Thomsen, University of Arkansas

Project description: Breakfast in the classroom (BIC) has been shown to meaningfully increase participation in the School Breakfast Program and reduce the number of schoolchildren who skip breakfast entirely. There is evidence that regular breakfast consumption positively impacts academic achievement and may be protective against excess weight gain. Thus, BIC may be an effective strategy to positively impact the lives of US schoolchildren. Existing studies on the effects of BIC are few in number and focus on children from large, urban centers. One concern is whether findings from these areas generalize to children in other parts of the country. To better understand the impact of BIC in reducing persistent disparities in academic performance and childhood obesity, we take advantage of the nation’s longest-running and most comprehensive statewide BMI screening program of public schoolchildren. These data reflect the population of Arkansas schoolchildren over a 16-year period. They will be linked to academic achievement outcomes and analyzed using strong quasi-experimental methods that include difference in differences estimation, matching, and synthetic controls. The study provides a novel opportunity fill a crucial knowledge gap by investigating the impact of BIC in less-urban contexts not adequately covered in earlier studies.

Did the Healthy Hunger-Free Kids Act help improve dietary quality among school-age children?

PI: Pourya Valizadeh, University of North Carolina at Chapel Hill

Project description: This study will assess the causal impacts of the 2010 HHFKA school meal nutrition standards on dietary quality of students participating in school meal programs with three specific aims. First, we estimate the average total/direct/spillover effects of HHFKA on the diet quality of children, overall and by school level (e.g., elementary vs. middle/high school), while accounting for endogenous program participation by directly modeling program participation within a simultaneous equation model. Second, we estimate the average effects of HHFKA on sub-categories of diet quality in a multiple programs- multiple outcomes framework. Third, we examine heterogeneities in the (total/direct/spillover) effects of HHFKA across the distribution of children’s overall diet quality. We use six waves of data from the continuous cycles of the National Health and Nutrition Examination Survey (NHANES), covering 2005 to 2016. Diet Quality is measured via Healthy Eating Index 2015 (HEI- 2015) and its components. We construct three HEI-2015 measures using food intakes from all sources, only school, and other sources than school to capture total/direct/spillover changes, respectively. Average effect analysis (aims 1 and 2) is conducted using a discrete factor simultaneous equation model. Distributional analysis (aim 3) utilizes a nonadditive fixed effects quantile regression model.

Does maternal depression caused by food insufficiency influence parenting practices and impact infant wellbeing? The role of SNAP.

PI: Irma Arteaga, University of Missouri

This proposal will examine the relationship among SNAP participation, maternal depression, and child’s health. We are specifically interested in understanding if SNAP would help parents to provide an environment that nurtures and fosters child’s health by working as a protective factor for maternal depression. In this study, we will advance the literature by using a quasi-experimental technique and a rich dataset. We propose to use siblings fixed-effects models and a unique dataset that combines a SNAP state administrative data with all Medicaid claims records associated with them. The advantage of using administrative data is that we know who participated in SNAP, how much they received, when they received it, and who entered and exited over time (self-reported data contains measurement error). The Medicaid claims data provides clinical diagnosis for postpartum depression and child’s health conditions. The large dataset will allow us to make comparisons between urban and rural low-income families, as well as between Caucasians and African Americans, something not yet examined in the literature. Consequently, this project has the potential to impact groups at higher risk for poor health by increasing our understanding of the associations between SNAP benefit and postpartum depression, well-child visits, immunizations, and child’s weight.

Food insecurity and child food consumption patterns among WIC participating families in Los Angeles County within the context of the WIC food package change and WIC + SNAP concurrent participation

PI: M. Pia Chaparro, Tulane University

Project description: The proposed study seeks to investigate, among WIC-participating families and those who concurrently participate in WIC and SNAP in Los Angeles County: 1) The impact of the 2009 WIC food package change on food insecurity prevalence and 2) the association between food insecurity and child food consumption patterns before and after the WIC food package change. Data from the triennial Los Angeles County WIC Surveys from 2005-2017 will be used for all analyses, including ~5000 randomly selected families who received WIC benefits in Los Angeles County the year the survey was conducted. In addition, information on food insecurity trends among eligible non-participants for both WIC and SNAP in Los Angeles will be obtained from the California Health Interview Surveys 2005-2017. Food insecurity was measured in the WIC surveys with a subset of questions from the U.S. Household Food Security Survey Module, whereas data on child food consumption includes number of servings of fruits and vegetables and of sugar- sweetened beverages, and frequency of fast food consumption. Multivariable linear and logistic regression models will be conducted as appropriate adjusting for child’s age, gender, and race/ethnicity; caregiver’s age, gender, race/ethnicity, educational attainment, nativity, and employment status; and family size, housing tenure, and income.

Labor supply distortions from nutrition assistance programs: Evidence from a bunching estimator

PI: Jason Cook, University of Pittsburgh

Project description: SNAP and WIC are primary components of the US safety net. Economic theory predicts that program eligibility and benefit formulas will reduce labor supply. Moreover, roughly 9 percent of SNAP recipients are able-bodied adults without dependent children and face strict work requirements, which may generate their own distortions. We provide the first evaluation of intensive margin labor supply distortions driven by key points of program formulas and work requirements. To do so, we exploit new administrative data from the Census Bureau on program participation for every SNAP and WIC recipient across 19 states that are matched to Census surveys and detailed earnings files from the Social Security Administration. These data allow us to explore these topics at an unprecedented level of breadth, depth, and accuracy. We will estimate distortions using difference-in-bunching designs that leverage the detailed nature of the administrative data in tandem with exogenous policy changes. This work is particularly policy-relevant considering the current efforts of the Trump administration to drastically expand work requirements in a host of safety net programs.

SNAP and work-related policies: An in-depth analysis of low-wage worker perspectives and behaviors.

PI: Caitlin Caspi, University of Minnesota

Project description: The study proposed is an in-depth analysis of low-wage workers related to SNAP and work-related policies. The aims are to: (1) to understand perspectives about current and future eligibility for SNAP benefits in two policy contexts (Minneapolis, MN and Raleigh, NC), and (2) to explore how, if at all, these perspectives affect decisions regarding household employment, major asset purchases, and savings. The study is embedded in a large, ongoing, NIH-funded natural experiment conducted in these two cities, in which 974 low-wage workers are approaching the second time point for data collection in a 5-year study to evaluate the health effects of an increase in the Minneapolis minimum wage. We propose to conduct 100 semi-structured interviews among workers who are either participating in SNAP, or who were participating in SNAP at baseline, but are no longer participating. This study offers a mixed-methods approach that will generate hypotheses about the mechanisms linking wage policies and the receipt of benefits, as well as help to interpret the findings of the larger study. This study is poised directly inform policy, specifically the design of minimum wage laws and the application of work-requirements for SNAP benefits, both on the horizon across the U.S.

SNAP, school meals, and the food security of multigenerational households.

PI: Agustina Laurito, University of Illinois at Chicago

Project description: The number of multigenerational households (with three generations or grandparents and grandchildren only) has been increasing in the U.S., especially since the Great Recession. There is evidence that these households are more likely to experience food insecurity, though this is not always the case. Additional SNAP benefits, as well as other assistance programs, may be one reason why multigenerational households are not always more food insecure. Yet, there is little work documenting the food consumption and food purchasing patterns of these households, and how they use different food assistance programs. This project seeks to fill that gap. Specifically, it investigates the interplay between SNAP and school meals (SBP and NSLP) and whether and to what extent this interaction helps multigenerational households maintain food consumption throughout the SNAP month. This project also helps illuminate intra-household dynamics and how those affect the food security of different household members (adults, seniors, adolescents, and children). It uses the National Household Food Acquisitions and Purchase Survey (FoodAPS); a nationally representative survey that collects detailed food acquisitions and nutritional information for all members of a household for seven days.

Understanding barriers to SNAP enrollment among college students.

PI: Maggie Dickinson, CUNY Guttman

Project description: SNAP participation rates among eligible college students appears to be much lower than average participation in the general population, despite high levels of food insecurity on college campuses. Eighty five percent of eligible Americans participated in SNAP in 2016[1]. While no national statistics on SNAP participation among college students exist, a recent report from the GAO indicates that at least 2 million eligible college students are not enrolled in the program [2]. The City University of New York, the nation’s largest and most diverse urban public university system, offers a natural experiment in understanding the causes of low enrollment among college students as well as the interventions that may boost enrollment of eligible students among this population over time. Drawing on a mixed methods approach, including surveys of students, focus groups, and interviews with staff engaged in SNAP enrollment on college campuses, this study will illuminate the factors contributing to low enrollment to inform meaningful interventions.