Date Approved


Date Posted


Degree Type

Open Access Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department or School

College of Technology

Committee Member

Dr. Yichun Xie, Ph.D. Chair

Committee Member

Dr. Daniel Fields, Ph.D.

Committee Member

Dr. John Dugger, Ph.D.

Committee Member

Dr. Alphonso Bellamy, Ph.D.

Committee Member

Dr. Huei Lee, Ph.D.


In the last few decades, the concept of a food desert has captured the attention of the public and academia. Research has referred to food desert as an area with lack of access to adequate healthy food, especially for those without a vehicle or who live more than ¼ mile walking distance from grocery stores. Although the food desert metaphor has been used in research in different disciplines such as regional planning and public health, this term has been used loosely without a precise definition and without identifying the factors that might impact it.

This research explores and examines the statistical association between demographic and socioeconomic characteristics and the existence of food deserts and access to unhealthy food outlets (e.g, fast food restaurants, convenience stores) in Wayne County, Michigan. This study obtained the precise location of healthy food outlets (e.g., grocery stores) and unhealthy food outlets (e.g., fast food restaurants) based on SIC codes, and U.S. Census data. The collected data were grouped into the distinct food oasis and food desert zones and analyzed using a GIS-network analysis, and exploratory spatial analysis, and a SPSS-statistical analysis. Another important goal of this research was to create statistical models to assess the effect of the selected explanatory variables such as race, level of income, and vehicle ownership on access to healthy food outlets in food desert and food oasis. In other words, these models can be used to find out how much the variables that have been selected can explain the type of food outlets, whether healthy or unhealthy, and how much variance in food oasis and food deserts can be explained by the selected variables. Finally, this study examined whether the absence of healthy food outlets leads to more geographical exposure to unhealthy food outlets.

This study has found that people of color and low-income populations are more likely than White and high-income populations to be located in food desert areas. However, ethnicity (Hispanic, Asian, and Arab) and gender (male, female) seemed to be less important in impacting access to grocery stores as an example of proxy to healthy outlets or fast food restaurants as an example of proxy to unhealthy food outlets based on our suggested regression models.