Author Identifier

Tony Greening

Date of Award


Document Type



Edith Cowan University

Degree Name

Doctor of Philosophy


School of Science

First Supervisor

Ute Mueller

Second Supervisor

Aiden Fisher


The spatialisation of violent crime is explored in two large case studies, Chicago and Sydney, using spatial econometric methods and macro-sociological variables derived from Social Disorganisation Theory.

Social Disorganisation Theory (SDT) is introduced in terms of its formulation in response to highly specific conditions arising in Chicago, as well as its adoption of methodological and theoretical developments from existing traditions. This specificity belies its breadth of application and enduring presence in criminology. With “Social Disorganisation Theory” hosting a wealth of highly nuanced academic dialogue conducted under its banner, current incarnations of SDT appear as branches on an evolutionary tree. This research addresses the theoretical roots of that tree, from which two primary benefits are derived. The first is that the resulting focus on macro-structural variables permits large-scale urban studies to be conducted with existing datasets. The second is that this effectively isolates the spatial analysis from specific theoretical developments and generalises the results. The difference from the classical formulation of SDT is that an “augmented” set of five variables is used as independent variables: disadvantage, population heterogeneity, residential mobility, family disruption and urbanisation. Criminal violence forms the dependent variable.

All variables are observed to exhibit spatial autocorrelation. In response, Spatial Durbin Models are selected for each case study. This selection is supported by diagnostics, with some qualification noted. Initial results suggest a basis for further exploration. In Sydney, this leads to a fully mediated model with family disruption as the mediator. In the case of Chicago, the strong landscape of segregation leads to a model which accommodates for the resulting structural instability. This introduces a model which provides separate treatment to highly homogeneous areas.

Results indicate mixed support for SDT. With the exception of heterogeneity and - to a lesser extent - urbanisation, variables broadly align with expectations derived from SDT in the initial Sydney and Chicago studies. However, these observations are muted by other outcomes. Firstly, the spatial complexity portrayed in the results is not formally conveyed by SDT. Using the argument that methodology both enables and constrains theory development, this is to be expected. Social Disorganisation Theory is credited with founding the ecological tradition in criminology. However, concepts of criminogenic place which have evolved from it are typically intertwined with co-location models of space. This is regarded as a limitation in which spatial autocorrelation is treated as a nuisance, rather than being theoretically embraced. That gap is highlighted by spatially rich results. The second threat to SDT in results from the Sydney case study is that a more parsimonious model is derived from family disruption alone. Furthermore, when disruption is employed as a mediator, full mediation is observed. The final response to SDT is that it does not accommodate structural instability as indicated in Chicago.

The results of this exploratory study offer insights into the spatial richness of violence in urban areas from an ecological perspective. This complexity poses a challenge to SDT. The thesis closes by discussing this challenge and includes an outline of proposed future work.

Access Note

Some figures and tables are not available in this version of the thesis due to copyright considerations.