Abstract
dc:description.abstractThis research examines the solvability factors of volume crime. It draws from a dataset of 836,616 robberies, violent crimes, sexual crimes, vehicle crimes and residential burglaries from the Metropolitan Police Service (MPS) area in London, England over two calendar years (2017-2018). It uses a process of inductive testing to identify likely solvability factors from the existing solvability literature as well as incorporating opportunities that exist because of how MPS data are coded. The solvability factors are reviewed, analysed and discussed by crime type. Solvability factors are initially grouped according to if they are binary (e.g., weapon used or not) ordinal (e.g., victim ethnicity), scalar (e.g., victim age) or if they are community factors (e.g., population density). After presenting results of inferential statistical tests (chi-squared tests) a Binary Logistic Regression (BLR) process is undertaken to review the solvability factors. Following this, an exploratory examination of the viability of a crime screening algorithm is undertaken for each crime type using the solvability factors identified. The purpose of this is to explore whether a subsequent, more rigorous approach, using solvability factors not available to this research, has the potential to improve the efficiency of the existing secondary investigation process, and is therefore worth taking to the next stage of development. The chapter that addresses each crime type concludes with a critical assessment of the ability of this research to add to the body of knowledge of solvability, the implications for Police practice and of the utility of the crime screening algorithm that was created. After the five crime types are discussed, there is a further chapter which addresses how the same solvability factors vary across the five crime types. This highlights how individual factors can vary, how their relative strengths differ, how they may operate in different directions in different contexts and the implications that arise from this. The thesis suggests that rather than a model of solvability being predicted by ‘victim status factors’ or fixed ‘situational factors’ or ‘police resources’, a new ‘pick and mix’ approach is needed. The thesis argues that a ‘pick and mix’ model of solvability should be built around core ‘tier-one’ factors (the most common, most powerful and most consistent solvability factors) and augmented by ‘tier-two’ factors (less common, less powerful, less consistent), that may vary by strength and direction of operation or be unique to specific crime types in specific jurisdictions. This thesis also suggests there is a good case to develop algorithmic supported decision making for the screening of two volume crime types (vehicle crime and residential burglary) to the next stage. This thesis then summarises where this research has added to the body of knowledge around solvability research and the implications for police practice. This added value arises from this research being conducted using one large dataset from one Force and the high-level access of the author. The new knowledge includes the identification of new solvability factors, the explaining of the variation of individual factors across different crime types, and the identification of several solvability factors (e.g., victim age) that are not simply linear in nature but apply in particular age ranges which also vary by crime type.
Degree
thesis:*- Name dc:type.qualificationname
- Doctor of Philosophy (PhD)
- Level dc:type.qualificationlevel
- Doctoral
- Grantor dc:publisher.institution
- University of Cambridge
- Year dc:date.issued
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Rose, Simon
- Advisor dc:contributor.advisor
-
- Campana, Paolo
Subjects
dc:subject × 8Rights
dc:rightsIdentifiers
dc:identifier.*- DOI dc:identifier.doi
- https://doi.org/10.17863/CAM.122798
- OAI identifier oai:identifier
- oai:www.repository.cam.ac.uk:1810/391806