In order to establish if criminals carried out distinct sub-sets of criminal activity as opposed to being widely versatile in the type of crime that they committed, 54 female and 49 male offenders completed a self-report questionnaire that asked about their criminality in two modes: Convictions and actual commission. MDS analysis of the resulting data supported the hypothesis that there is a subset of general criminal activities that most offenders have been involved in, but that there is also a tendency for offenders to evince thematic foci to their activities that relate to dishonesty, violence or antisocial crimes. These three core themes are present both in male and female offenders and across data modalities, emerging in the crimes for which the offender has been convicted as well as those to which s/he admits.
In order to reveal the variations that exist among stalkers in terms of actual stalking behaviours the range of offence actions that should be examined was derived from theoretical accounts of the differences between offenders. Four distinct thematic foci were hypothesised characterised by sexual, intimate, possessive and aggressive-destructive modes of offender-victim interaction. To test these hypotheses 50 offences were content analysed into 24 behavioural categories. The occurrence of these categories of behaviour across all offences was examined using SSA-I. A modulating facet was proposed, by analogy with previous studies, of violent sexual assaults that reflected the intensity of personal contact the stalker imposes upon the victim, with the most intense being the most differentiated and least frequent. The results lend support to the existence of a intensity facet that modulates all four aspects identified from the published literature, providing further evidence for a radex of criminality. This radex model is used to indicate biases in the current sample of cases by postulating implicit elements for future study. The implication of the radex model of stalking for the management of and interventions in stalking are also considered as the basis for future explorations.
Two contrasting hypotheses about criminal organisations are that they will be either strongly hierarchical with recognisable differentiation between components of their organisational structure or that they will be loose networks of associates with no discernable organisation. Six indices of organisational morphology were therefore derived from Social Network Analysis measures to explore the different organisational structures of criminal groups and networks. These were used to classify 12 drug-dealer networks, 11 property crime networks and six hooligan networks. Partial Order Scalogram analysis of these 29 groups across the six measures showed that there were a wide range of group structures, ranging from very loose networks without even central key figures through to highly structured networks that contained all six indicators of structure. Within this framework two dominant axes were identified, along which the degree of structure varied. One was proposed as a product of the size of the group, the other as a product of the centrality of leadership. This allowed three types of criminal organisation to be specified, A-ad hoc groups, B-oligarchies and, C-organised criminals. These different types tended to relate to group size with A as the smallest and C the largest with on average almost twice as many members. These different types of group carry different implications for group process and investigative interventions. Although there was a tendency for the hooligan groups to be less structured and the drug networks to be the most structured, most forms of structure were found for all types of criminal activity. The study therefore raises a number of important questions about what gives rise to these variations in organisational morphology and the implications the variations have for the conduct of the crimes, their investigation and methods for reducing the effectiveness of criminal networks.
In this study, the process of burglars' specialisation was examined. 15 sub-categories of burglaries committed by 3,066 burglars were analysed, using a thematic approach. The result of an SSA-I showed that four themes existed in the structure of burglary: ‘residential’, ‘commercial’, ‘public’, and ‘industrial/storage’, Also it was found that ‘residential’ and ‘commercial’ burglaries were the most distinct from each other, providing for two dominant foci for burglaries. The results of POSA also confirmed that most burglars specialised in either ‘residential’ or ‘commercial’ burglaries. Few were specialised in ‘public’ or ‘industrial/storage’ burglaries. In addition, the number of people who specialised in commercial burglaries decreased with the increase of burgling experiences. The psychological processes underlying burglars' specialisation in one theme, or shift from one to another, are discussed. This study shows that the thematic approach offers a method of studying the multidimensional nature of burglars' psychological processes of specialisation.
Exploratory Spatial Data Analysis (ESDA) is applied to the spatial pattern of residential burglaries in Tokyo, which was caused by new lock-breaking tools, called “lock-picking”. The official crime records of residential burglary and breaking and entering in the 23 wards in Tokyo from 1996 to 2000 (n = 41151) are aggregated by census tract and then matched with the national census data in 1997. The burglary rates are calculated by house type: detached, high-rise and low-rise housings. The burglary rate increased only at high-rise housings after the new lock-breaking tools appeared. Global and local Moran's I statistics revealed that the clusters of detached house burglary were most significant and static, while those of high-rise housing displaced dynamically. Other findings are reported and discussed.