Changing types of homicide in Scotland and their relationship to types of wider violence
Item statusRestricted Access
Embargo end date04/07/2019
Skott Bengtsson, Anna Sara Viveka
The lack of information about the relationship between homicide and violence was identified as a gap in knowledge almost 30 years ago. Despite this, little research has been conducted worldwide regarding this relationship on a national level since then, and the results of that research have been very contradictory. This lack of research includes Scotland, despite its unenviable reputation of being the most violent country in the Western world. Even so, many studies make unsupported assumptions regarding the relationship between the trends in homicide and wider violence. In order to fill this gap in research, the aim of the thesis is therefore to examine the changing characteristics and patterns of homicide in Scotland and to determine the extent to which changes in homicide reflect the changing characteristics and patterns in wider violence. Overall, both homicide and violence have more than halved over the past twenty years in Scotland. But this is not just a numbers game. Due to the heterogenous nature of these crimes, although the overall picture is one of decline, there might be certain types of homicide and violence that have remained stable, or even increased over this time. In order to examine the relationship between homicide and violence in Scotland, subtypes of both homicide and violence were identified and compared over time. Two datasets were used in the current study; a homicide dataset gathered from the Scottish Homicide Database, spanning from 1990-2015, and a violence dataset gathered from pooled survey sweeps of the Scottish Crime and Justice Survey, spanning from 2008- 09 to 2014-15. Multilevel latent class analysis was used to identify subtypes of both homicide and violence using classifying variables relating to the victim, offender and to the incident of lethal and non-lethal violence. This study presents the first use of this type of multilevel latent class analysis in all criminological research. The results identified four main types of homicide (Stabbing homicides, No Weapon-bludgeoning homicides, Rivalry homicides and Femicides) and four main types of violence (Domestic, Public No Weapon, Public Weapon, and Work-related). When the homicide typology and the violence typology were compared over time it was found that although there are some differences in the subtypes identified, the overall trends in these two crimes seem to follow a similar pattern over time. A key finding from this study is that the general decrease in both homicide and violence was driven by a reduction in the same type of violence, namely violence committed by young men in public places and involving the use of sharp instruments. However, this general decrease in violence masks a hidden relative increase in both lethal and non-lethal forms of domestic violence over time. This thesis will argue that the trends in homicide and violence indeed do follow a similar pattern over time, but that an overall picture of decline does not mean that all types of violence or homicide are decreasing equally. This has vital implications for violence policy. Improved and specific prevention strategies are needed for certain types of lethal and non-lethal violence, such as domestic violence, in order to ensure that all types of violence are prevented equally. This study will also make important theoretical contributions, in that all theories making assumptions about the trends in homicide and violence should examine disaggregated subtypes of these crimes in order to provide a holistic explanation of the changes in these crimes. Limitations of the study are discussed as well as future implications of these findings for policy and theory.