This thesis investigated the distribution of community-based crime prevention programs across 223 suburbs in Brisbane, Australia. It aimed to problematise the implementation and funding of services and examine whether the number and type of programs implemented were related to crime rates, community capacity, community perceptions of collective efficacy or other community characteristics.
The thesis addressed two research questions. First, what types of community-based crime prevention programs are most prevalent in the Brisbane Statistical Division? Secondly, how are resident demographics, community cohesion, community levels of collective efficacy and community crime rates related to the distribution of community-based crime prevention programs across the Brisbane Statistical Division? Community-based crime prevention programs were defined as those that might impact one or more crime risk factors. This broad definition includes not just programs targeting crime and harm prevention, but also programs with aims such as safeguarding against social isolation and providing welfare support. I used four main data sources. The first was an audit I conducted of all community-based crime prevention programs in Greater Brisbane, active at some point from 2000 to 2005 inclusive. I found 8,894 programs, funded by a variety of organisations, and coded the crime prevention approach of each: 1. problem-focused, 2. developmental (modifying individual pathways), or 3. social-community (strengthening the community).
The second data source was the Australian Research Council Community Capacity Survey: Wave One, a survey of 82 randomly selected communities in Greater Brisbane. This survey captured each community’s perceptions of their levels of cohesion, disorder, social activism, and collective efficacy.
The third dataset consisted of community characteristics and demographics from the Australian Bureau of Statistics (ABS) 2001 Census. The fourth dataset was Queensland Police Service data on annual crime rates, by crime type and community.
I examined the theoretical construct of collective efficacy to determine whether there was a relationship between program density and a community’s belief in a common sense of trust and cohesion and willingness to solve local problems. I found no statistically significant relationship. Interestingly, program density instead had a strong positive correlation with civic engagement – the extent to which people involve themselves in activities for their community’s benefit. Due to this correlation with civic engagement, but not collective efficacy, I propose that it is not residents initiating an intervention, but residents participating in an intervention, that impacts program density. Organisations can analyse the needs of a community and develop and organise interventions, but to keep a program running it is necessary to engage community residents willing to participate in those interventions.
By comparison, I investigated whether there was a link between low program density and residents’ fear of participating. My research supported this hypothesis, finding an inverse relationship between program density and resident perceptions of crime levels. Specifically, program density had moderate or strong inverse relationships with community perceptions of community violence, alcohol and drug problems, community problems and community divisions.
The relationships between program density and actual crime rates were not strong, but were positive, the opposite to the relationship of program density with perceptions of crime. Program density had a moderate positive relationship with property crime and weak positive relationships with drug, violent, and public nuisance crimes.
In summary, this thesis had two major findings. The first finding was that it was not the perceived level of collective efficacy but the level of civic engagement in a community that had a significant relationship with program density. The second finding was that the mix of program types did not differ greatly across communities; this was potentially due to programs being implemented without investigation of community needs, as well as being implemented broadly. I argue that it is important to understand that programs are not implemented in isolation; communities often already have multiple co-existing crime prevention and other programs. A regularly updated and searchable database of existing programs, like the one constructed for this thesis, would be a useful resource for both government and communities interested in developing and evaluating programs.