This thesis is an investigation into support for terrorism and political violence in the UK. It has two principal aims: to advance a new application of rational choice theory capable of providing an explanation of support for violence that accommodates the impact of religion, and to test it using quantitative data.Drawing on existing work from criminology and political science, it hypothesizes that amongst those with a political grievance violence will be supported most often where the benefits of democratic political engagement are viewed unfavourably. More than this, it argues that regular exposure to those holding extreme religious views may deplete political trust, and thus lower the favourability of these perceptions, thereby increasing the likelihood of an individual supporting violence.Two distinct approaches were adopted to test this theory. The first employed multivariate analysis of secondary data, drawing on the European Values Study (EVS) and the Ethnic Minority British Election Study (EMBES) to test for a relationship between attitudes to democratic political engagement and support for violence in the British population at large, and to explore the impact of religion on these attitudes amongst a sample of British Muslims. The second approach sought to build on these findings, using data from an experiment conducted over the social networking site Facebook to explore whether attitudes towards democratic political engagement could be manipulated to trigger a shift in support for violence.The results of these analyses offered limited support for my theory, with outputs from both the EVS and EMBES data showing a significant relationship between attitudes to democratic political engagement and support for violence, despite the failure of the experiment to find evidence supportive of a causal relationship. Analysis of the EMBES data similarly affirmed a role for regular exposure to those holding extreme religious reviews in predicting both attitudes to democratic politics and support for violence, though this was shown to be non-linear. The implications of these findings for existing scholarship on - and policy responses to - extremism and support for violence were considered, alongside the limitations of my approach. The opportunities for future research were then presented alongside a summation of my key findings.