Find a SECReT supervisor
prism apply now
Download SECReT Brochure
News and events
@UCLCrimeScience Mailing list

SECReT student seminars 2012

Unlocking the investment returns of effective crime reduction programmes: why particular interventions work, and how they can be implemented effectively in the UK context

Publication date:

Start: Oct 30, 2012 12:00:00 AM

Jacqueline Mallender, CEO Matrix Knowledge

Governments across the world invest in interventions to promote law and order, prevent crime and reduce offending.  In recent years, international research has evolved which demonstrates the effectiveness, or otherwise, of these interventions (see for example http://www.campbellcollaboration.org/coordinating_groups/index.php).  Current and recent Departmental and Cabinet Office initiatives demonstrate that UK policy makers are keen to develop and share knowledge in this field and to understand why particular interventions work, and how they can be implemented effectively in the UK context (see for example the recent discussion with stakeholders about “what works centres”).. Increasingly foundations (such as the Monument Trust) are also looking at effectiveness when considering funding applications from third sector innovators seeking to develop new interventions in this field. As outcome-based contracting develops, commissioners and providers of for correctional services in the UK will need to ensure that evidence of effectiveness underpins service delivery agreements and associated financing. Interventions also cost money. Policy makers are increasingly looking beyond the impact on crime and re-offending and are asking about value for money, affordability, and funding priorities. Initiatives such as the Social Finance Bond are being promoted by government as a means of unlocking the investment returns of effective crime reduction programmes and enable effective interventions to receive much needed seed funding. The lecture will provide an introduction to this issue and how economic methods in this area are being developed. Some case studies will be presented to show economic analysis of crime interventions and crime programmes will be presented.

Bio: Jacqueline is CEO of Matrix Knowledge and is an international health and welfare economist with over 30 years’ experience. She started her career as a government economist in the UK, firstly in the Department of Health and subsequently as the Health Economic Advisor in H.M. Treasury. Jacqueline then worked as an economist with Coopers and Lybrand before leaving to found MHA a specialist health economics consultancy working in the NHS and which was the foundation of Matrix Knowledge (TMKG Ltd). During her career she has worked in fields of health, crime and justice and education in the UK and internationally. She is a convenor of the joint Campbell and Cochrane Economics Methods Group and is on the Steering Committee of the Campbell Collaboration Crime and Justice Group.

Diagnosing and preventing corruption

Publication date:

Start: Sep 20, 2012 12:00:00 AM

Professor Adam Graycar, Australian National University (ANU)

Adam Graycar is Professor of Public Policy at the Australian National University (ANU), where he is also Director of the Research School of Social Sciences. He joined ANU in 2010 when he became the Foundation Dean of the Australian National Institute for Public Policy for two years.  He has had long experience in both academia and in government.  His most recent government position was Head, Cabinet Office, Government of South Australia, and his most recent academic post before joining ANU was Dean, School of Criminal Justice, Rutgers, the State University of New Jersey.  He also served as Director of the Australian Institute of Criminology for nine years.

Adam’s current research is on corruption, and he is the Director of the Transnational Research Institute on Corruption at the ANU.

He has worked in many policy areas and in many policy settings.  He has had long experience in policy making, research, and research management at the most senior levels in Australia and internationally.  He has run two major national social research agencies.  He has acquired extensive policy experience over 22 years in the senior level posts he has held in government in Australia, both Federal and State.  He has two doctorates from the University of NSW, is the author of some 200 scholarly publications, in fields such as political science, public administration, sociology, social policy population studies, criminology.

Crime Patterns and Spatial Choice: Theories, Models and Some Evidence

Publication date:

Start: Jul 16, 2012 12:00:00 AM

Wim Bernasco, The Netherlands Institute for the Study of Crime and Law Enforcement (NSCR)

Most behavior of interest to social scientists is choice behavior: deciding to take a course of action while forgoing alternative options. Criminology is no exception. This presentation deals with the spatial choices that offenders make: where to go, how far to travel, in what direction, and how? In geographical and environmental criminology, these questions have traditionally been studied in research traditions isolated from each other. Journey-to-crime research, for example, explores how far offenders travel to crime, but has not really investigated where they go. Ecological studies of variations in crime rates have emphasized the role of attractive criminal opportunities, but they have ignored that offenders must know about such opportunities in order to profit from them. Recently, the spatial discrete choice framework was introduced in geographic criminology. It explains the offender’s choice of where to commit an offense, and it integrates the study of spatial crime distributions with journey to crime research. Developed in micro-economics, it is firmly rooted in principles of rational choice and utility maximization, and requires input from more substantive theories, such as crime pattern analysis from criminology, learning theory from psychology, or foraging theory from behavioral ecology. A number of recent applications address criminal location choice in burglary, robbery and other crimes. They have demonstrated that the spatial discrete choice framework can successfully describe spatial decisions as a function of characteristics of the target locations and the characteristics of the offender. In particular, the findings show that offenders are attracted to places that are accessible to them (e.g. nearby their homes), where the expected profits of crime are high while the risks of apprehension are low. Other applications have established that social barriers inhibit the journey to crime, that highways can deter short journeys to crime but facilitate longer journeys to crime, and that former anchor points (past homes) play a role in offenders’ crime location choices. Like all questions pertaining to geography, the spatial discrete choice framework is saddled with the challenge of defining the appropriate spatial unit of analysis. New developments in the area of crime location choice include a focus on small spatial units of analysis (such as street segments), the assessment of spatial spillover effects, repeat crime location choice, and crime location choices that are conditional on time.


Dr Wim Bernasco is a senior researcher at the NSCR.  His current work focuses on spatial aspects of criminal activities, including variations in crime and delinquency between neighbourhoods, offender travel behaviour and target selection.  He was the first researcher to employ McFadden’s discrete choice approach in the study of where offenders choose to offend and has published several seminal works on this topic. 

How Cryptosystems Are Really Broken

Publication date:

Start: Jun 20, 2012 12:00:00 AM

Prof Adi Shamir, Weizmann Institute

Most of the cryptosystems we currently use are highly secure, and cannot be broken by mathematical cryptanalysis. However, over the last fifteen years researchers have developed many types of physical attacks on their implementations which can easily bypass their mathematical security. In this talk Prof Shamir surveyed some of the latest attacks, and showed how difficult it is to build a truly secure communication systems.

Sex, race and offending trajectories: An analysis of an Australian longitudinal offending database

Publication date:

Start: Jun 18, 2012 12:00:00 AM

Professor Anna Stewart, Griffith University

Anna Stewart is a professor in the School of Criminology and Criminal Justice at Griffith University. From 2008 – 2010 she was the Head, School of Criminology and Criminal Justice.  Prior to this in 2007-2008 she was the Deputy Dean (Learning and Teaching) in the Faculty of Humanities and Social Sciences.  In 1994 she received her PhD from University of Queensland in psychology. The topic of her thesis was An investigation of decision making by child protection workers. Since this time she has been interested in the research uses of government administrative data.  Using these data she has built two longitudinal birth cohort databases, linking information about individuals born in 1983/1984 and 1990 from child protection, youth offending and adult offending databases.  Her current work involves linking these data to data from Queensland Health regarding system contacts for mental health concerns.  She has examined the lifetime contacts individuals have with child protection, youth justice, adult criminal justice systems, system responses to youth offending and domestic violence, management of risk, diversionary responses and system modelling. She has published over 60 peer-reviewed publications, government reports and non peer reviewed publications. She has been involved in partnerships that have obtained over four million dollars in National Competitive Funding, consultancies and other government research funding.  Her work receives strong support from across Queensland Government criminal justice agencies.

Forensic Computing - A Beginners Guide

Publication date:

Start: May 9, 2012 12:00:00 AM

Denis Edgar-Nevill, Founding Chair, British Computer Society Cybercrime Forensics Specialist Group

A Scientific Investigation of Blast Injuries: London 7/7 Terrorist Bombings

Publication date:

Start: May 8, 2012 12:00:00 AM

Dr Hasu Patel, Consultant Plastic and Reconstructive Surgeon at St Bartholomew’s and the Royal London Hospitals

The UK National Risk Assessment

Publication date:

Start: May 2, 2012 12:00:00 AM

Anita Friend, Assistant Director Risk Assessment, Cabinet Office

Anita Friend presented on the UK’s National Risk Assessment and the National Security Strategy.

The Strategies of Kidnappers: Understanding violence during kidnapping for ransom negotiations

Publication date:

Start: Apr 19, 2012 12:00:00 AM

Dr. Everard Phillips

Research suggests that kidnappers who ransom their hostages use violence towards their hostage in calculated or sometimes ritualistic manner in order to coerce the target of the ransom. Yet, little is empirical analysis
of these methods and how these actions constitute a crime commission process. This seminar briefly introduced two studies that examine how hostages are coerced in captivity or prepared for the eventual ‘proof of
life’ process. In addition too, examining how this behaviour reflects the types of tactics that kidnappers use to negotiate a ransom demand. In particular, how kidnappers use violence towards their hostage during the
hostage negotiation, as well as the strategic implication of these actions. These factors will relate to understanding whether the hostage will be killed by their captors.

Dr Everard Phillips has been researching the strategic behaviour of kidnappers who ransom for over 10 years. A Criminal Psychologist Everard specialises in profiling factors that lead to post-settlement hostage
homicide, hostage victimisation, and the tactical negotiation behaviour of kidnappers who ransom. Although his focus covers contemporary kidnapping for ransom activity as seen in Iraq and Afghanistan in recent years, his
primary focus is behaviours of the differing organised criminal groups that operate in Latin America, South East Asia and Africa. Everard has spoken extensively on the subject to a broad spectrum of audiences that have
included diplomatic, military, various international risk and security organisations.

Illicit activity in prisons - how can technology help?

Publication date:

Start: Apr 4, 2012 12:00:00 AM

Dr Mireille Levy, NOMS Science and Technology Advisor
Martin Lee, Head of the NOMS Intelligence and Operations Unit

Reducing illicit activity in prisons is a constant battle in which technology has a huge role to play. The talk introduced the National Offender Management Service (NOMS) and gave an overview of the main criminal threats inside prisons and of current and future mitigation strategies. It looked in more detail at drugs and the concrete problem of reducing supply and use in a diverse and continuously evolving environment. The talk reviewed the technologies currently available to detect drugs in prisons and set out the challenges for further research.

Dr Mireille Levy

After a distinguished research career as a physicist at the Rutherford Appleton Laboratory, Dr Mireille Levy joined the Home Office in 2001 to manage police radios. She is now the Science and
Technology Advisor for the National Offender Management Service. Her remit covers all science and technology issues relevant to prison and probation, from preventing use of mobile phones and drugs inside jails to equipping prison staff with modern telecommunication systems and developing the next generation of electronic tagging. She is a Member of the Institution of Engineering and Technology and a Fellow of the Institute of Mathematics and its Applications.

Martin Lee

By profession I am an analytical chemist specialising in the drugs field having worked for many years  in the Forensic Science Service as a forensic toxicologist.I have worked extensively with law enforcement agencies on all aspects of policing, including a spell as a senior adviser to HM Chief Inspector of Constabulary. More recently I lead for NOMS on the prisons drug strategy, which included the commissioning of drug treatment and drug testing services and the oversight of our
research programme. As lately head of the Intelligence and Operations Unit, I had lead policy responsibility for the key threats faced by NOMS. I am a Chartered Chemist,a Member of the Royal Society of Chemistry and a Fellow of the Royal Society of Arts.

Spatial is Special: Interdisciplinary Research at CASA

Publication date:

Start: Feb 22, 2012 12:00:00 AM

James Cheshire, UCL Centre for Advanced Spatial Analysis

Everything happens somewhere: this presentation outlined some of the ways in which researchers at the UCL Centre for Advanced Spatial Analysis (CASA) are collecting, analysing and visualising unprecedented volumes of spatial data. At CASA, these data have been applied to a range of studies, from mapping peoples’ surnames to their commutes on public transport; from movements during London’s Riots to linkages between our online and offline behaviour. Such research has many applications beyond its “traditional” audience of planners and geographers and I hope that this talk will generate some new perspectives and collaborations from the SeCRET research community.

Strategic security planning for the built environment

Publication date:

Start: Feb 29, 2012 12:00:00 AM

Colin McKinnon, Innovation Director, Buro Happold Ltd
Andrew Sieradzki, Director of Happold Safe and Secure.

Buro Happold is a professional services firm providing engineering consultancy, design, planning, project management and consulting services for all aspects of buildings, infrastructure and the environment, with its head office in Bath, Somerset.[5] Originally working mainly on projects in the Middle East, the firm now operates worldwide and in almost all areas of engineering for the built environment, with offices in seven countries. The parent company owns the subsidiary companies such as Happold Safe and Secure. The firm includes a number of specialist engineering consultancy groups, including fire engineering and lighting consultancy.

Buro Happold talked about how they designs security into some of their major projects using examples such as the Emirates Stadium. They also talked about some of their research and projects.

Evolving the Face of a Criminal

Publication date:

Start: Jan 31, 2012 12:00:00 AM

charlie frowd

Dr. Charlie Frowd, University of Central Lancashire.

EvoFIT is a software system that helps victims and witnesses of crime generate an image of a criminal’s face. It draws on Dr Frowd’s research looking at techniques to improve the effectiveness of facial composites, so that eyewitnesses can build a clearer, more accurate image of the person they saw commit a crime. The pioneering system requires witnesses and victims to repeatedly select from screens of complete faces, with ‘breeding’, to ‘evolve’ a composite of the criminal’s face. It is currently used in crime investigations throughout the world because of the high identification rate of its composites.

EvoFIT composites have been made even more effective with a novel animation technique that Dr Frowd has developed.  The technique progressively caricatures a composite, similar to the way sketch artists work, exaggerating characteristic parts of the face.  Police can use this image format when making public appeals for information on TV and online newspapers, substantially increasing the chances of identifying the criminal.

Background to the work

Dr Frowd’s research explored the effectiveness of traditional methods for constructing composites using software programs in current police use. Such systems build a face by witnesses and victims selecting individual facial parts: hair, eyes, nose, mouth, etc. The research has found a startling result: a composite is rarely named correctly when constructed a couple of days after the face had been seen, the norm for witnesses and victims of crime. As a consequence, and part of collaborative research, Charlie has initiated and led a range of projects that have sought to improve the quality and recognition of these images. He has also about 40 research papers published in the field.

Charlie’s PhD in Stirling involved development of a new composite system. It is called EvoFIT and was inspired by how we perceive human faces, as whole entities, and also by selection and breeding processes that are found in nature. EvoFIT presents users with sets of complete faces and they
select a few that resemble the criminal. The software then breeds these faces together to produce more faces. While the faces at the start have random characteristics, repeating the selection and breeding process a few times allows the system to converge on a specific identity. Using
procedures that mirror policework as far as possible, the latest version of EvoFIT now evolves a face that can be correctly named 45% of the time when constructed following a two day delay. EvoFIT is in use in a dozen police forces across the UK and abroad, and reports from the police indicate a
success (arrest) rate in the region of 50%.

While EvoFIT has led to significant improvements in facial identification, it has always been difficult to get a good-enough image to reproduce in newsprint. However, Dr Frowd’s team have experimented with “stretching” the faces and getting people to look at them sideways on, rather than looking straight at the image in front of them.

In testing, people were twice as likely to correctly identify a composite when it was stretched and turned sideways (36%) as opposed to when they viewed them normally (17%).

Dr Charlie Frowd is based in the School of Psychology at the University of Central Lancashire, Preston. He started his academic career twelve years ago at the University of Stirling, in Scotland, where he completed his PhD in Psychology and then worked on a series of government-funded grants with Vicki Bruce and Peter Hancock. Two years ago, he joined the School of Psychology at UCLan and is now a senior lecturer. His main research interest is in the area of police facial composites (pictures of suspects to crime, as seen in the newspapers and on TV crime programmes).

The Development of a Wireless Electrostatic Mark Lifting Method and its use at Crime Scenes

Publication date:

Start: Feb 15, 2012 12:00:00 AM

The Development of a Wireless Electrostatic Mark Lifting Method and its use at Crime Scenes
Robert Milne and Grahame Sandling, CSI Ltd

CSI Ltd have two areas of interest. The first is to do with materials involved in a technique called electrostatic mark lifting, which was demonstrated during the seminar.  The second involves pattern recognition.  See presentation below..

CSI Ltd produce their own footwear evidence software called Treadmark, which currently has the latest UK beta version working very well in the Leicestershire Constabulary which has the best footwear evidence detection rate in the UK. The software stores footwear marks and scans of suspected offenders shoes for measurement comparison and overlay enabling the ease of handling of the evidence type for screening. The software connects on line to the National Police Improvement Agency's (NPIA) National Footwear Reference Collection (NFRC) of over 19,000 reference records from anywhere on the internet. We began to look at machine vision and Treadmark can for zigzag patterns search on wavelength, amplitude and periodicity but we seek to improve on this basic approach and see a considerable market in effective automatic pattern recognition not only in our application but in many other areas.

CSI Ltd has produced a students (Colleges) version of Treadmark (one licence for 5 stand alone computers) in modern 64 bit format ready for market so that students of forensic science can actually use software actually used by police services and access the NFRC. The stand alone licence version can however communicate with other stand alone's or Treadmark in networks by prior arrangement. In this way small police services can afford a system and be networked without the expense of actually having a server based licence designed for larger organizations or states.

Search UCL News