Innovations in Quantitative Methods PhD Opportunity:
Agent-Based Modelling of Crime
Centre for Applied Spatial Analysis and Policy
Centre for Criminal Justice Studies,
University of Leeds, West Yorkshire, UK
Crime is one of the most significant influences on people living in contemporary societies; from the simple act of locking our doors through to the complex systems of law that constrain our behaviour, crime is a key element of all our lives. Understanding the processes and drivers behind crime is important for developing effective crime prevention strategies. Environmental criminology has highlighted the individual-level drivers of crime and focuses on the effects of the environment surrounding a crime event and the behaviour of the people involved. On the other hand, crime data, which drives the financing of anti-crime agencies and is significant in the development of anti-crime policies, is generally at the aggregate level, removed from individual level behaviours. There is still a fundamental lack of understanding as to how individual-level behaviour dynamically plays out to create larger patterns of crime and, in reverse, how crime distributions reflect individual behaviour. Modelling crime across these scales allows us to estimate how policy changes will play out, and to run experiments of crime prevention measures - thus connecting individual-level research and policy-driving data to the advantage of both fundamental research and crime-prevention.
Agent-based modelling is a technique of increasing popularity and high potential that models systems by simulating the behaviour of the individual components of the system (e.g. burglars, victims, households) directly, rather than modelling the system at an aggregate level (as, for example, statistical regression does). This PhD will focus on the development of the SimBurglar model - one of the most advanced agent-based model of crime in existence. The PhD will extend the model with novel work on the daily/seasonal behaviour of potential victims, along with a more sophisticated representation of criminals informed by qualitative behavioural profiling. These developments will be founded on a study of the spatial and temporal patterns of crime within Leeds, UK, which will be used to elucidate the interaction between specific types of victim and offender. This study will inform the model, which, in turn, will enhance our ability to predict crime hotspots and test whether government crime-prevention and urban regeneration policies will have an effect. It will also allow us to understand current crime trends better, utilising, enhancing, and critiquing current qualitative crime theory.
Applications are invited from candidates from either computational science or social science backgrounds; candidates with a good understanding of criminology or socio-economics are welcome, and training in programming and agent-based modelling can be provided - though a strong desire to engage in programming and modelling social systems is a pre-requisite, experience is not. Typical subjects applicants may have studied include computer science, criminology, geography, sociology, psychology, as well as any degree involving the analysis of complex dynamic systems (for example, applied mathematics, physics, or statistics). This PhD will allow the student to develop skills suitable for careers in high-level policy planning, crime and socio-economic research, agent-based modelling, and complex systems analysis, as well as academia. Opportunities to apply for international collaboration are currently available and encouraged through the World Universities Network's Researcher Mobility Programme.
The award will cover fees, maintenance and research training at standard ESRC rates. Currently this is £13,590 tax-fee maintenance per year, plus £750 per year research budget, plus fees. The studentship is available either for three years as a full time PhD student for students with a suitable Masters or skillset, or for these three years plus an additional year's funding for an initial Masters course (MSc in Geographical Information Systems) if the student doesn't yet have a Masters in a cognate subject area and needs to develop their quantitative skills (for details of whether your Masters/skillset qualifies, please contact Dr Evans, below).
Deadlines for applications: 29th April 2011
Project supervisors:
Dr Andrew Evans, Senior Lecture in Geocomputation
http://www.geog.leeds.ac.uk/people/a.evans/
Prof. Susanne Karstedt, Professor of Criminology and Criminal Justice
http://www.law.leeds.ac.uk/about/staff/karstedt.php
Dr Nick Malleson, Research Fellow in e-Infastructure for the Social Sciences
http://www.geog.leeds.ac.uk/people/n.malleson
Please contact Dr Evans ([log in to unmask]) for further information.
-----------------------------
Professor David S. Wall
Criminology, SASS
Durham University
32 Old Elvet
Durham
DH1 3HN
Email [log in to unmask]
Tel: 0191 334 6983 (direct)
Tel: 0191 334 6827 (general office)
|