This thesis investigates issues of home ownership, life satisfaction and criminal activity and their relationship to each other. Drawing from the body of economic theory and econometric techniques, currently available, it explores the extent to which these topics, significant concerns for society, and individuals within society, can be examined. A variety of statistical techniques, from a range of disciplines, alongside contributions from alternative literatures, are utilised to discuss the subject. Four empirical chapters, connected to the areas above, consider recent publications, and offer distinct approaches to investigating the subject. Chapter one considers the importance of housing tenure and quality in terms of life satisfaction. Central to the argument is the view of Dietz and Haurin (2003), supported by a range of disciplines, that the individual life satisfaction benefits enjoyed by homeowners are attached to the pride and self-efficacy derived from becoming owners, and that these gains are directly connected to housing quality concerns. This chapter reports on analysis made through the use of a fuzzy measure of housing quality within a Mundlak corrected panel ordered probit. Results suggest that including a control variable for housing concerns mitigates the relative life satisfaction gains enjoyed bV homeowners. Chapter two builds on the analysis presented, by examining the importance of homeownership and its relationship to fear of crime. van Praag, and Ferrer-i-Carbonell (2008) demonstrate a novel approach to modelling ordered dependent variables which allows for a multi-level linear approach with instrumented variables, referred to as probit adapted ordinary least squares (POLS). This cross sectional approach identified the significance of individual crimes and fear as a whole in determining life satisfaction. Chapter three draws upon recently published work by Cook and Winfield (2013) to model convergence levels with previously unconsidered data and with a focus on the importance of the urban rural divide in criminal activity. With use of Beta and Sigma convergence findings suggested that, in the US, smaller counties experienced less convergence, and that the level of disaggregation was important for viewing the emergence of a national crime trend. Chapter four expands on the notion of disaggregation by investigating the presence of the motivation and opportunity effects between crime and economic conditions, recently identified by Cook and Watson (2013). The chapter draws upon the works of Phillips and Land (2013) whilst expanding the data sets used and offering additional empirical approaches. The repurposing of the Pesaran and Timmermann (1992) test of forecast accuracy is used, together with additional cross section and time series approaches, in order to test the relationship originally proposed by Cantor and land (1985). Results find that viewing state and county level disaggregated data offered a relationship in contrast to those found by national research.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:644532 |
| Date | January 2014 |
| Creators | Winfield, Thomas George |
| Publisher | Swansea University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0014 seconds