ANALYSING AND MODELLING CRIME AND PLACE Chair: Christophe Vandeviver Chair: Christophe Vandeviver

management of compliance in England and Wales. Thus, the paper will contribute to our understanding of both the ‘relationship’ and compliance - two important aims and mechanisms of probation practice. Using emotional labour in the context of probation will also allow us to think about sources of burnout, stress, and low morale. We will end the paper by drawing on existing literature to think about the implications of the Transforming Rehabilitation agenda on the way in which probation practice can be seen as a form of

‘emotional labour’ and what might lie in store for practice as changes to the structure of probation in England and Wales are implemented.

3.12 ANALYSING AND MODELLING CRIME AND PLACE

179 0231 - DO AREA CHARACTERISTICS INFLUENCE THE LIKELIHOOD OF OFFENDING? AN INVESTIGATION USING STRUCTURAL EQUATION MODELLING

Jack Cunliffe (United Kingdom)¹ 1 - London School of Economics

Diverting an adolescent from beginning an offending career is better than subsequent punishment or incarceration. In short, prevention is better than cure. We are all familiar with the phrase ‘growing up on the wrong side of the tracks’, but capturing the effect of area on offending behaviour has proved difficult.

This work, conducted as a PhD, has two main aims. The first is substantive: to investigate whether and how an individual’s perceptions of their area act as risk factors for offending.

Area effects have historically been under researched, in part due to a lack of appropriate data but also due to methodological limitations. The second is methodological: to demonstrate that theoretically-informed, structural equation modelling can make best use of existing and often under-utilised datasets, particularly cross-national studies such as those typically conducted by large scale organisations or governments.

Using the United Kingdom Offending, Crime and Justice Survey (OCJS) conducted between 2003 and 2006, and taking a range questions on individual perceptions, family circumstance, self-reported offending and variables relating to the area in which the respondent lives, the work reviews previous criminological measurement constructs of well-known risk factors (from both an analytical and theoretical perspective) and once these are defined moves onto structurally modelling self-report offending.

Findings show that the individual perceptions matter most, with a tangled interrelationship of area perceptions operating in conflicting directions. Once this is accounted for, living in an area with higher disorder seems to slightly increase self-report offending with part of the relationship explained by lower collective efficacy. Limitations, both of the data and the analytical technique will be discussed and the implications for other works and data sources touched upon.

0232 - MODELING RESIDENTIAL BURGLARS' TARGET SELECTION PROCESS AT THE HOUSE-LEVEL

Christophe Vandeviver (Belgium)¹

1 - Institute for International Research on Criminal Policy (IRCP), Ghent University

What residence attributes affect burglars’ target choice and why do some burglars target remote residences? This study taps into rational choice theory to first investigate how burglars select a target by relying on residence attributes to optimize a combination of perceived rewards, efforts and risk, and then tests the hypothesis that higher perceived rewards, lower

effort and lower anticipated risk compensate travel effort. Using data on 650 residential burglaries committed by 650 unique burglars during the period 2006-2012 and the approximately 500,000 residences in one Belgian province in a discrete spatial choice framework with the residence as the spatial unit of analysis, we find that burglars rely on effort-related attributes to distinguish between targets. Higher perceived rewards decrease the odds of a residence being burglarized. Risk-related attributes are unimportant for burglars’

target choice in general. With regard to selecting remote targets, the results show that lower risk compensates travel efforts and that burglars aim to make minimal effort, even when they are already confronted with increased travel efforts. No support is found for the hypothesis that higher perceived rewards compensate increased travel efforts.

0233 - AWARENESS SPACE AND DISTANCE DECAY BUFFERS: CONTRADICTION OR CONSEQUENCE

Stijn Van Daele (Belgium)¹ 1 - Ghent University

Offenders’ awareness space (Brantingham & Brantingham, 1981) is believed to be one of the central concepts that shape geographic offending patterns, as it is assumed that offenders operate in areas they are familiar with. Offending patterns themselves empirically result in a distance decay curve: most crimes are committed close to offender’s residence and gradually decline as distance increases. Although awareness space makes no reference to this, the distance decay curve does not demonstrate straightforward decay when the smallest distances are considered: it appears to contain a buffer very close to the offender’s home, where little to no offences are committed. This is assumed to be caused by offenders avoiding the higher risks of recognition near their residence (Rossmo, 2000). Given the lack of this assumption in the whole awareness space concept and because of the strong link between awareness space and distance decay, two possibilities can be considered: either the Brantinghams’ awareness space lacks a necessary concept – and is therefore a too parsimonious concept in its present form – or the buffer is not necessarily the result of recognition avoidance – and Rossmo’s assumption is therefore not parsimonious. The capacity of awareness space to generate a buffered distance decay curve is tested in an agent-based model. Using a basic representation of the awareness space an simple offending behaviour rules in an agent-based model, we find that such a buffer can be explained without making assumptions about recognition avoidance and that a buffer effect may occur as a mere consequence of a simple conception of awareness space.

181 3.13 VIGNETTE EXPERIMENTS IN CRIMINOLOGY

Chair: Stefanie Eifler

0234 - VALIDITY ASPECTS OF VIGNETTE EXPERIMENTS Stefanie Eifler (Germany)¹

1 - Catholic University Eichstaett-Ingolstadt

This presentation gives an extensive overview of applications of vignette experiments in criminology. It covers their methodological advantages as well as their disadvantages. In particular, the question of validity aspects is addressed – the problem of survey error due to social desirability and the problem of closeness of vignette experiments to actual situations.

0235 - SANCTIONING EFFORTS DURING CONFLICTS - AN APPLICATION OF THE SCENARIO-TECHNIQUE

Lena Verneuer (Germany)¹ 1 - University of Bielefeld

The panel study Crime in the modern City [Crimoc] focuses on the emergence and development of deviant and delinquent behaviour of adolescents resp. young adults. Since 2002, self-report data is collected by a standardized questionnaire (paper-pencil method) and thus carries all characteristics of a typical quantitative survey. In 2013, a vignette, which was set up for measuring different (hypothetical) reactions to deviant behaviour according to a conflictual situation in a discotheque, was (re-)implemented in the questionnaire.

In the context of this talk, both the scenario-technique in the Crimoc-Project and the application of this scenario are brought into focus. Applying an additional scenario and combining it with the average survey-data enables one to describe reactions to deviant behaviour in a conflictual situation with far more precision. The question to what extend these reactions can be understood as sanctioning efforts due to a violated sense of justice is analysed with the given (panel-)data.

Crimoc-Project: www.crimoc.org

0236 - “DON’T BLOW YOUR COOL” – SELF-CONTROL AS A RESOURCE FOR COPING WITH PROVOCATION

Sonja Schulz (Germany)¹

1 - GESIS – Leibniz Institute for the Social Sciences

General Strain Theory has received broad empirical support, but little is known about potential moderators of the strain-delinquency relationship (Agnew 2013). This study tests whether self-control attenuates the relationship between a certain type of delinquency – violence – and its most important precursor, which is considered a type of strain: interpersonal provocation.

Applying two different scenarios designs, the process is examined using randomly varied degrees of objective provocation as well as a measure of subjective sensitivity to provocation.

Drawing on recent work which suggests differentiating between the personality traits of risk-affinity and impulsivity, this study compares the conditioning effects of both characteristics on the provocation-violence link. The analyses are based on a large sample of seventh-graders (n

= 2,635) from five cities in Western Germany interviewed in the classroom in 2013. Linear probability models regressing past violent behaviour and prospective violence on provocation measures, personality traits, and interactions between provocation and both traits are estimated. Findings underscore that self-control indeed functions as a coping resource that enables students to control their anger and to refrain from translating aggressive impulses into action. This study further highlights the importance of using internal consistent and mechanism-congruent measures in the study of illicit coping processes and conditioning factors and discourages from using composite, potentially multidimensional measures.

Agnew, Robert. 2013. "When Criminal Coping Is Likely: An Extension of General Strain Theory."

Deviant Behavior 34:653-670.

3.14 PROPERTY CRIMES: OPPORTUNITY AND CONTROL