Research

Published Papers

Schauder, Stephanie A., and Mark C. Foley. “The Relationship Between Active Transportation and Health.” Journal of Transport and Health 2, no. 3 (2015): 343-349.

Active transportation has received attention for its environmental and health benefits. In this paper we use individual-level survey data from NHANES III to investigate the extent to which the number of minutes of bicycling and walking for transportation are associated with 10 health outcomes. We use instrumental variables to address the endogeneity caused by the complex relationship between exercise and health. We find mixed results indicating that active transportation is associated with improvements in some health outcomes such as weight and cholesterol, but not others such as systolic blood pressure and glycohemoglobin.

Other Papers The Effect of Sprawl Development on Grocery Store Location and Food Access(Job Market Paper)

This work explores the effect of sprawl on food access in the United States. The development of the interstate highway system since the 1950’s reduced transportation costs and many cities developed in a sprawling pattern characterized by low population density and car dependence. This paper seeks to understand how sprawl development affects car dependence. I use an agent-based model to generate a testable hypothesis. Then I use an instrumental variables (IV) model to test this hypothesis with data from 239 major U.S. cities. 1947 planned highway rays is the instrument. The simulation suggests that urban sprawl, characterized by car dependence, reduces the number of grocery stores per population. The IV model tests this hypothesis and finds that cities with greater sprawl development and car dependence are more likely to have food access problems. Specifically, a 1 percentage point reduction in the number of people driving to work alone is associated with a 1.4 percentage point reduction in the percentage of census tracts with low access to food, and a 1 percentage point increase in the number of people who walk reduces the percentage of tracts with low food access by 5.2 percentage points.

“Income Segregation and Access to Healthy Food,” with Shyam Gouri Suresh*

We build an agent-based model to test the effect of heterogeneous transportation costs and income segregation on grocery store location in a linear city.  We draw inspiration from and build off of the literature of the Hoteling model.  Agents are individuals and grocery stores.  Individuals are given an income and location and maximize their utility function which contains healthy and unhealthy food as well as transportation effort.  Individuals are assigned to drive or walk which affects transportation effort and is correlated with income.  Firms maximize profit and we use best response dynamics to find each firm’s optimal strategy which includes a location and path for prices.  We vary income segregation to observe how grocery store location and pricing change in response.  We find that increasing income segregation, causes low-income agents to be farther from a grocery store and high-income agents to be closer to a grocery store. We also find that nutritious food is less available to low-income individuals when our model incorporates economies of scale.

“The Effect of the Medicaid Expansion on Opioid Addiction”*

In this paper, I examine the effect of the Medicaid expansion on opioid addiction.  I hypothesize that expanding Medicaid coverage could potentially improve indicators of opioid addiction if people are able to enter addiction treatment.  Alternatively, expanding insurance could worsen the epidemic if it permits more people to seek medical assistance for their pain and opioids are covered by insurance.  As indicators of opioid addiction, I look at the effect of Medicaid on deaths due to opioid overdose, opioid addiction treatment admissions, and opioids sold.

“Use of an Agent-Based Model for Insight into the Role of the Fresh Fruit and Vegetable Program on Preference Formation” with Michael Thomsen and Rodolfo Nayga*

The Fresh Fruits and Vegetables Program (FFVP) is an initiative started by the U.S. Department of Agriculture to improve healthy eating among children.  The FFVP program provides a free fresh fruit or vegetable snack to children in qualifying schools up to five times a week.  It is important to evaluate this program and understand its effects on childhood nutrition.  We build and agent-based model of preference formation to understand how exposure to fresh fruits and vegetables in early elementary school affects healthy food consumption by 6th grade.  We use the temporal difference learning algorithm popularized by Hammond et al. (2012).  This model simulates habit formation in the context food by modeling preference formation a dopamine signals that give the brain positive or negative feedback about the food most recently consumed.  We use data from Arkansas schools, to look at how different patterns of exposure to FFVP affect preferences, and how living in a food desert changes the effect of FFVP.  We find that early exposure is more beneficial than late exposure to FFVP conditional on the number of years exposed.  We also find that FFVP is more beneficial for children who grow up in environments lacking healthy food because they have the most to gain from the intervention.  This study can help policy makers understand the optimal way to distribute limited FFVP funding. 

*Please contact me at sas648@cornell.edu for a copy of any of these papers

Research Assistant Work

Researcher for Jean-François Houde Cornell University | 2018 I used Stata to analyze data from a randomized control trial on a desludging project in Senegal. I created regression tables, figures, and I contributed to the writing of the paper.

Researcher for Lars Vilhuber                                                                    Cornell University | 2017 I worked with the LDI research group at Cornell University to use and improve U.S. Census Bureau data. I created code in R to impute the county variables for the Synthetic Longitudinal Business Survey.

Researcher for Young Jo and Brandon Restrepo                     USDA Economic Research Service | 2018 I worked at the U.S. Department of Agriculture Economic Research Service on a project involving the Early Childhood Longitudinal Study (ECLS-K). We used Stata to ascertain the age that racial disparities in weight emerge and to shed light on potential causes.