M. Leitner1, J. Kent2, J. Austin3

1 - Louisiana State University, Department of Geography and Anthropology, Baton Rouge, USA

2 - Louisiana Geographic Information Center, Louisiana State University, Baton Rouge, USA

3 - The JFA Institute, Washington DC, USA


The natural disaster that took place on August 29, 2005 (Hurricanes Katrina and Rita) in Louisiana has created a historic opportunity to evaluate our understanding of the impact of massive population displacement and resettlement on crime, community and the criminal justice system.


This research assesses the impact of massive displacement of people (non-criminals and criminals) on crime rates both within Louisiana and selected urban areas outside of Louisiana (Houston, Atlanta). Within Louisiana certain parish populations have dramatically increased in a relatively short period of time, including Baton Rouge, Shreveport, and many rural parishes.


Three different types of data are used in this analysis. As denominator data to calculating crime rates, population counts were collected from the US Census Bureau and from applicants for the FEMA Individual Assistance (IA) program. Reported crime data used in this research represent the seven index-crimes that define Part-I criminal offences: murder, rape, robbery, aggravate assault, burglary, larceny theft, and motor vehicle theft.  Part-I offences were provided by the Louisiana Commission on Law Enforcement for the years spanning 2000-2006.

The analysis methods are centered on a Geographic Information System (GIS), which is used to input, store, analyze and map all data (population, crime, socio-economic). Three different types of analysis are conducted. First, exploratory spatial data analysis (ESDA) is conducted with all data. This includes summary statistics in tables, graphs, and maps and provides a first visual impression about changes in crime rates and their relationships with data known to be associated with crime (population, socio-economic). The interaction of space and time is analyzed with a rather new technique called the co-map. The co-map builds upon a technique known as the co-plot that is used to examine the relationship between a pair of variables (in our case the x and y locations of a polygon centroid that includes a crime rate) that is conditioned using a third variable (z in our case a measure of time, such as month or year). A second type of analysis is statistical and focuses on the detection and mapping of high concentrations of crime (i.e., crime hot spots) and their changes over time. The third type of analysis is also statistical. It uses different spatial regression models to analyze the relationships (and their changes) between crime (dependent variable) and population and other demographic and socio-economic data (independent variables) believed to be associated with crime.