Strengthening the effectiveness of ‘whole of government’ interventions to break the cycle of violence in disadvantaged communities

The vision of burning cars, conflicts between police and local residents, and strident calls for increased law and order intervention on talk-back radio remind us that crime, violence, perceptions of a lack of safety and security and marginalisation are the everyday experiences of many socially disadvantaged communities in NSW. While Redfern, Macquarie Fields and Dubbo have recently made the headlines, similar problems have been reported in Claymore, Miller and Bourke and other parts of NSW in the last decade.

The underlying problems of these communities are often endemic and related to a lack of opportunities to fully participate in society.¹ The nature of the problems are well known to residents of the area, local service providers and local government, as well as state and national parliamentarians. Within these communities there is a general acceptance that ‘tough on crime’ strategies will be only part of action to prevent the conditions in which violence arises. There is general acceptance that a ‘whole of government or community’ approach is required that increases the visibility of police and access to human services, improves the amenity of local areas and builds the capacity of the local community to identify and work towards solving problems.² Because these ‘whole of government’ approaches can be resource-intensive and challenge organisations to work in new ways, there is increased pressure for evidence of effectiveness before these approaches are mainstreamed into routine ways of working.³

This paper is based on learning gained through the evaluation of ‘two whole of government’ interventions in communities in western Sydney. Both of these interventions arose in response to violent acts within these communities that resulted in injury and extensive property damage. These events mobilised local politicians, government and non-government service providers and citizen groups to come together to work on ‘joined up’ solutions. There was a wide range of interventions from increased police presence in the area, improvements in lighting and street scapes, outreach and co-location of human services, community development programs and housing improvements.

As evaluators of these programs, we became aware that there was often a ‘program logic’ gap between what was agreed as the rationale for the intervention, the expected outcomes and the actions undertaken. For example, it is often stated that the active engagement of the community in defining problems and seeking solutions is essential for long-term gains. However, in looking at project documentation and actions it was often unclear how this engagement would occur, especially in a context where timelines and project deliverables needed to be developed with short time frames before the programs had begun.

In their work on realistic evaluation, Pawson and Tilley state that we need to be clear about the overall goals of the intervention(s), the mechanisms or pathways through which they will be achieved, the expected outcomes and the contextual factors that will make this set of interventions unique.⁴ As many of these interventions have not been researched and reported in the peer-reviewed literature, the failure to have a clearly articulated program logic adds to the difficulty of building a systematic evidence base of the effectiveness of ‘whole of government’ programs. Difficulties in developing program logic can be compounded when people are drawn from cross-disciplinary teams and professional groups that have differing views of the most effective approaches to common problems and ways of assessing evidence of effectiveness.

Based on our experience in evaluating the impact of ‘whole of government’ approaches in these communities we have identified two areas where fundamental differences in perspective between partners can occur and, if these are not recognised, make it difficult to develop an agreed program logic. These are: ‘what do we mean by community?’ and ‘what are we trying to change or influence?’.

We have found it useful to think of a community in three ways, as this understanding affects how people think about intervention.

• A venue: a place where interventions can easily reach a large number of people;

• A setting: a defined organisation or structure where it is possible to work with people living and working in the area to achieve a set of agreed goals; and

• A dynamic system: a complex web of structures and relationships, webbed into the external environment, where action requires an internal and external focus that is constantly evolving.^5,6

Box 1 provides an example of how these different interpretations of community influence responses to an issue.

We have found that people involved in whole of ‘government’ initiatives often have difficulty in linking what they are trying to change with what they are doing and what they are expecting to achieve. For example, building a youth centre may be expected to have an impact on school retention rates, vandalism and depression. However, an examination of the strategies to be undertaken and the numbers of young people attending the youth centre suggests that these outcomes are unlikely to be achieved.

One way of unpacking these links in a community context is to decide if the focus of the intervention is:

• The people who are living in the community, where the focus is on changing specific individuals’ life chances or addressing specific problems, such as getting young people back to school;

• The sense of place in the community, which McIntyre describes as the sociocultural and historical characteristics of communities, including shared values, traditions, norms and interests;⁷ or

• The space within the community, such as the quality of the buildings, physical infrastructure, and natural environment.⁸

The concepts of ‘space’ and ‘place’ are often used interchangeably but their distinction is essential for focussed intervention. While ‘space’ describes where a location is, ‘place’ describes what a location is. Place is, therefore, an interpretation of physical space and community and has been linked to constructs of identity, illness and disability.⁹ Box 2 provides an illustrative example.

Complex, inter-related problems require complex intervention. Few of us are working in environments where we can undertake extensive planning and review processes when responding to a community crisis that brings with it opportunities for the funding of new programs and ways of working. But we do need to develop an evidence base that demonstrates that what we are doing has the potential to make a difference.

At a practice level this can be done through systematic development of program logic to drive the interventions. Wherever possible this should be informed by the use of research evidence and subjected to rigorous evaluation. This can be facilitated by:

• Building a consensus across several different organisations about what we hope to achieve. This needs to recognise that each organisation will have their own ‘core business’ which affects how they see the problems and develops priorities for action. For example, some organisations may have a much stronger commitment to promoting access to services as a solution to problems, while others may have a stronger commitment to community development approaches.

• Recognising that the people working across the many organisations involved will be drawn from different disciplines and professional groups and that these backgrounds will inform the ways in which problems and solutions are defined and understood.

• Identifying what is already known about what works by drawing on the experiences of many different disciplines, including criminology, public health, political science, community psychology, sociology, human geography and community development.

• Training workers from a range of organisations and disciplines on evidence of effective action across these disciplines.

• Providing opportunities for reflection on areas where there are high levels of agreement and those where consensus is an ongoing issue.

Grappling with these sometimes difficult but important issues is essential if we are to build a shared and pragmatic evidence base for whole of government approaches in breaking the cycle of violence in disadvantaged communities.