Academic Research

Publications

"Food Choices at a Client Choice Food Pantry: Do Low-Income Pantry Users Respond to Changed Opportunity Costs?" (with Tammy Leonard, UT Southwestern Medical Center)

Food Policy

Abstract: Client choice food pantries allow individuals, many of whom are food insecure, to select a preferred bundle of food. To date, interventions to improve the nutrition of food choices in pantries have not included price incentive programs like those employed in the retail food sector because pantries do not charge for food. However, economic incentives may still play a role in food pantry choices through choice architecture. We examined a natural experiment involving two client-choice regimes that effectively altered the opportunity cost of food selections. Longitudinal individual fixed effects models provide evidence that pantry clients responded to changed opportunity costs by selecting more foods that became relatively less expensive and fewer foods that became relatively more costly. Our study highlights the impact of choice architecture, and in particular relative trade-offs, on food selections in the food pantry context. 

Working Papers

Does DACA Affect Labor Market Outcomes? Evidence using a Cross-Sectional Differences in Regression Discontinuity Design (sole author)

Abstract: This paper studies the impact of DACA on DACA eligible individuals using a regression discontinuity design (RDD) and a cross-sectional differences-in-regression-discontinuity (DRD) design. Using data from the American Community Survey (ACS), I find that DACA positively impacted the income of the DACA eligible population. 

INDUSTRY/CONSULTING RESEARCH

“Revenue Analysis of a Major US retailer: ‘Subs by Tubs’,” (with Robert Peyton Santori and Sonja Hightower) – (This project won the first position at Buxton Co., Fort Worth)

This research analyzed the factors that influence the profits of a major US retail chain called “Subs by Tubs.” After completing the pre-model analysis of cleaning the data set and dealing with missing observations and outliers, several different regression models were estimated using the OLS procedure. The independent variables for the model were chosen using a combination of theory, logic, statistical significance, and model scoring. After the “best” model was selected, predicted sales were calculated for various potential locations. Subsequently, recommendations were given to the Buxton Company regarding the performance of potential future locations and whether or not each potential location was deemed to have a “high revenue potential.”