Use this URL to cite or link to this record in EThOS:
Title: An environmental and behavioural analysis of arson in a Danish sample
Author: Kappel, C.
ISNI:       0000 0004 7658 142X
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Background: Despite its significance as a costly and destructive criminal behaviour, there appears to be some consensus that we know relatively little about arson compared to many other areas of criminal behaviour. Moreover, most existing theory and research into arson has come from the USA, and has tended to concentrate on profiling the characteristics of offenders, rather than investigating, at least in any detail, features of the environment that may influence their behaviour. Aim: The main aim and overarching theme of the current thesis was, therefore, to describe and evaluate some of the main demographic and biographical characteristics, offence related behaviours, and environmental factors associated with arson, in a sample of arson offenders from a European country. Methods: Six empirical studies were conducted, each based on cases drawn from a sample of 746 cases committed by 540 offenders from Denmark between 2002 and 2010 in two police districts, one rural and one urban. Studies 1 and 2 examined a range of demographic and biographical characteristics of arson offenders (such as, gender and age); Studies 3 and 4 covered offence related behaviours (such as selection of targets, and travel distances), and employed regression analyses to look specifically at how these were predicted by other offence related and demographic and biographical variables. Study 5 then investigated the prediction of serial offending as an indicator of arson recidivism using the above demographic and biographical variables and offence related variables. Finally, Study 6 attempted to employ a new approach, via Google Earth, to examine the influence of a range of architectural and structural features of the environment on arson offending; these included 2 targets, presence of high rise buildings, territorial markers, population density and maintenance. Results and Discussion: Findings supported previous literature in identifying the typical arsonist as a young male offender; however, the results further suggested three possible divergent trends in the data corresponding to different categories of arson offender: 1) a more frequent opportunistic arsonist; 2) a less frequent, but more serious, often more persistent serial offender, and 3) a category of mainly female offenders who are less likely to be serial offenders but who may be reacting to dysfunctional home environments. Importantly, in relation to the latter finding, a bimodal peak in age emerged in the subgroup of female offenders, identifying a younger group of female offenders in their mid and late teen years and an older subgroup of female offenders in their late thirties and early forties. Another notable finding was that young male offenders who were not at school were particularly at risk for becoming serial offenders, suggesting that young males not attending school could be targeted in terms of prevention of persistent arson. Also, as predictors of arson, a number of environmental variables were significant and in line with predictions (for example, arson was more prevalant where there were vacant buildings, and very significantly, where the nearest police station was farthest away), but others were significant in a direction opposite to predictions (high building density was associated with lower rates of arson), and some potentially important predicted relationships failed to emerge as significant predictors (such as territorial markers). In addition to the above, two other major findings emerged. First, whilst it was possible be to predict crime scene behaviours from other crime scene behaviours with some degree of accuracy, and, similarly, demographic behaviours (like previous arson) from other demographic factors, predicting crime scene behaviours from demographic factors and vice versa proved to be considerably more difficult. In contrast, in terms of having 3 maximum impact on arson rates the environmental variables considered here did a relatively good job of predicting the presence of arson. A number of limitations and implications are also discussed. Conclusion Considering the results as a whole, notwithstanding some success in predicting arson from demographic and offence related variables, it is concluded that an extension of the kind of environmental approach explored in this thesis could potentially be used for developing environmental schemes for arson prevention that might be considerably easier to apply, and perhaps even more effective in reducing arson, than targeting 'at risk' groups of individuals.
Supervisor: Wagstaff, Graham Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral