Date of Graduation


Document Type


Degree Name

Master of Science in Agricultural Economics (MS)

Degree Level



Agricultural Economics and Agribusiness


Rodolfo M. Nayga

Committee Member

Michael Thomsen

Second Committee Member

Bruce L. Dixon


Social sciences, Arkansas


In this paper we utilized a panel data set from 2004 to 2010 to identify and determine the demographic and economic drivers of food deserts in both urban and rural areas in Arkansas. We defined food deserts as areas where access to healthy foods such as fresh vegetables and fruits are limited. More specifically, separate distance measures from the census block centroid to the nearest supermarket or grocery store were used to determine if the area is an urban food desert (1 mile) or rural food desert (10 miles). These distance measures were then aggregated at the census block group level. Locations of supermarkets and big grocery stores that provide fresh produce were geocoded (latitude and longitude) accordingly. Socio-demographic and economic variables at the census block group level were then matched with the distance information. These variables were from Census 2000 Summary File 3. Finally, we employed multivariate regression approaches to model the relationship between socio-demographic and economic factors and the existence of urban and rural food deserts in Arkansas. We found that block groups with deprived situation, such as less per capita income, higher unemployment, and less educational attainment, will be more likely to be food deserts.