Negative perceptions of anti-social behaviour have been shown by previous research to have harmful repercussions to both an individual‘s mental and physical health as well as the neighbourhood‘s long term prospects. To track such perceptions the Labour administration of 1997 to 2010 developed quantitative measures at both the national level (using the Crime Survey for England and Wales until recently known as the British Crime Survey) and at the Local Authority level via Place Surveys – a postal survey of residents in all Local Authorities. This thesis argues that the Place Surveys were methodologically flawed. Multilevel small area synthetic estimation can provide an alternative to such localised surveys by using statistical models that predict the probability of a target variable using national data, but adjusting that prediction to take account of local characteristics of both the place itself as well as the people living there. The overarching aim of this thesis is therefore to provide, for the first time, a truly localised picture of perceptions of alcohol and drug-related anti-social behaviour across all English neighbourhoods. The validation tests on the synthetic estimates calculated for this thesis demonstrate a high degree of concordance with the vastly more expensive Place Surveys thus showing that synthetic estimation of crime and criminal justice issues based on the Crime Survey can add value for money to existing datasets, as opposed to spending substantial sums of money on poorly answered local surveys.
|Date of Award
|Liz Twigg (Supervisor), John Mohan (Supervisor) & Catherine Jones (Supervisor)