Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

Legacy Department

Civil Engineering

Committee Member

Dr. Wayne Sarasua, Committee Co-Chair

Committee Member

Dr. Jennifer Ogle, Committee Co-Chair

Committee Member

Dr. Ronnie Chowdhury, Committee Member

Committee Member

Prof. Stephen Sperry, Committee Member


This research was conducted to assist in transportation safety planning at both a macro (statewide) and micro (neighborhood) level of geography. Addressing safety issues at high crash incidence locations through crash countermeasures or better geometric design helps to make our roadways safer; however, the most influential and ever-present factor in most crashes, the human factor, is not directly addressed. Therefore, the primary goal of this research was to identify and analyze phenomena about the residences of drivers involved in crashes using spatial and statistical methods. These phenomena include socioeconomic and demographic characteristics of neighborhoods where these drivers live and the proximity of crashes to driver residences. Understanding the correlation between the densities of drivers involved in crashes and characteristics of neighborhoods where they live may help to optimize expenditure of scarce safety funds on safety programs that better target current and future high risk drivers. To add to this goal, a more focused probe into young driver behavior was done through an investigation into teen driver crash involvement within South Carolina public high school districts. Also, an investigation into the proximity of traffic crashes from driver residences was done to identify any relationships or possible correlations with trip lengths. The residential locations of drivers involved in crashes in South Carolina, found using 9-digit zip codes acquired from the South Carolina Department of Motor Vehicles (SCDMV), were crucial to the success of this research. Other important data elements needed for this research were: spatially accurate crash data, census socio-demographic data and boundaries at a relatively fine scale (block group level), high school attendance zone statistics and boundaries, routable street networks, and statewide grid cells at one square mile resolution. A combination of spatial analysis techniques (crash location coordinate geocoding, driver residential 9-digit zip code geocoding, block group aggregation, cluster analysis, grid cell aggregation, and network analysis) and statistical analysis methods (odds ratio, risk ratio, correlation analysis, multiple linear regression and Chi-square tests) were used in this research. The results of the spatial and statistical analyses conducted in this research demonstrate the significance of relationships between high and low density clusters of drivers involved in fatal and injury crashes (at-risk drivers) and the socio-economic and demographic characteristics of the residential areas where these at-risk drivers live. For example, the median household income and educational attainment (at least college degree attained) variables showed a negative correlation to the at-risk driver clusters, meaning that areas with high median household income and high educational attainment were more likely to have fewer at-risk drivers than other areas. Also, the regression estimates suggest that public high school zones with high graduation rates, high overall enrollment, and less money spent per student (low poverty index) have a lower rate of young driver involvement in fatal and injury crashes compared to zones with low graduation rates, low enrollment, and more dollars spent per student (high poverty index). Although the proximity analysis results suggest that approximately 35% of crashes occur within 5 miles of the driver’s residence, the risk ratio analysis shows that considering only trip length, the probability of being involved in a fatal or injury crash is lower for trips closer to home when normalized based on the number of actual trip lengths. Lastly, the one square mile grid aggregation of both at-risk drivers and crashes of specific contributing factors help zero in on areas to be concentrated on from a safety program implementation and enforcement standpoint. This research could potentially assist the decisions of state officials with regard to selecting and implementing transportation safety programs and strategies for the safety emphasis areas in South Carolina’s current strategic highway safety plan, ‘Target Zero’. Overall, a more holistic approach to transportation safety would be to encourage transportation professionals and state policy makers to consider the approach taken in this research where drivers are made the focus of transportation safety in combination with the more traditional methods of addressing safety through countermeasure implementation and better geometric design, thus optimizing the use of limited state funds and resources.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.