Title page for ETD etd-10142009-154753


Type of Document Dissertation
Author Kent, Joshua D.
URN etd-10142009-154753
Title Essays on the Integration of Anisotropic Landscapes within Contemporary Geographic Profiling Models
Degree Doctor of Philosophy (Ph.D.)
Department Geography & Anthropology
Advisory Committee
Advisor Name Title
Leitner, Michael Committee Chair
Braud, DeWitt Committee Member
Lam, Nina Committee Member
Rohli, Robert Committee Member
Weindorf, David Dean's Representative
Keywords
  • spatial analysis
  • environmental criminology
  • serial crime
  • journey-to-crime
  • Bayesian journey-to-crime
Date of Defense 2009-09-16
Availability restricted
Abstract
A criminal geographic profile is a tool used by law enforcement professionals for estimating the probable location of a serial offender’s anchor point, or domicile. This estimate is constructed according to the distribution of linked crime scenes. While this approach can utilize a number of different modeling techniques, most fail to account for the inherent irregularities of the physical and cultural landscape. Contemporary methodologies have consistently adopted the a priori assumption that an offender’s crime scenes and anchor point are located across an isotropic surface on which the opportunity to offend is equally distributed around the offender’s residence. Simple introspection clearly reveals that this assumption is unrealistic. Landscapes are comprised of heterogeneous collections of physical and cultural features that, by their very nature, bias the location of human activities and the occurrence of phenomena. Accordingly, this research examines four alternative techniques for geographically profiling offender behavior in space using models that are capable of accounting for the intrinsic irregularities present within the surrounding landscapes. The first technique examines the efficacy of functional distance metrics for interpreting an offender's perceptions of travel cost. The second method estimates the serial offender's anchor point by modeling the spatial variability observed for the linked crime scenes using a dynamic standard deviational ellipse. The next method extends the premise of a non-uniform landscape by introducing land cover characteristics within a probability distribution strategy. Finally, a land cover enhanced profiling technique is proposed using an empirical Bayesian formulation. Comparative analyses of the four enhanced techniques validate the premise that landscapes impart a deterministic impact on a serial offender's behavior in space. Consequently, these factors can be incorporated within various analytical frameworks to produce an accurate and precise estimate of an offender's anchor point.
Files
  Filename       Size       Approximate Download Time (Hours:Minutes:Seconds) 
 
 28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access 
[LSU] kentdiss.pdf 5.02 Mb 00:23:15 00:11:57 00:10:27 00:05:13 00:00:26
[LSU] indicates that a file or directory is accessible from the LSU campus network only.

Browse All Available ETDs by ( Author | Department )

If you have questions or technical problems, please Contact LSU-ETD Support.