Skip to content

Download as PDF [1MB]

Chapter Six: Research

Key Points

There is an urgent need to improve the evidence base to inform the delivery of high-quality services to people vulnerable to homelessness. Since the pool of money is limited, it is important that funded programs are cost-effective and work to protect and enhance the life chances of people who are homeless or at risk of homelessness. People who are homeless or are at risk of homelessness are among the most vulnerable in our community - it is vital that interventions actually work and that they do no further harm.

Rigorous, accurate and reliable national data across the service systems and program areas working with people who are homeless or at risk is needed to reliably measure progress in addressing homelessness. Better information is needed on the many different pathways people take through the service system to measure longer term social and economic participation outcomes and to improve the service offer to people who are homeless or at risk of homelessness.

Strategies to reduce homelessness should be informed by research, critical evaluation, practitioner expertise and the needs of individual clients of specialist homelessness services. Progress made towards the goals and targets set in the White Paper will need to be measured to ensure accountability of all levels of government.

Existing data on homelessness

Census data: Counting the Homeless

The Australian Bureau of Statistics (ABS) has developed its Homelessness Enumeration Strategy since the 1996 Census to undertake a rigorous count of the homeless population.  This strategy was refined for the 2001 Census, with data collectors working with state and territory governments and homelessness providers to ensure an accurate count.  Detailed analysis of this data set was undertaken and a report on the homeless population, with an estimate of the number of people who are homeless, was made. 

Using Census data to measure progress against strategies to reduce homelessness presents a number of difficulties.  These include the challenge of counting populations such as rough sleepers, the inability to capture longitudinal data on individual pathways in and out of homelessness and the frequency of the collection.  A further difficulty is the lag time between data collection and availability of the data analysis—data from the 2006 Census was not publicly available until 2008.

While the Census data provides a snapshot of the homeless population on the particular day that the Census is taken, it does not capture the duration of homelessness or the reasons for homelessness.  Nor does it provide information about who is at risk of homelessness. 

In an attempt to gain more frequent and comprehensive data about people who are homeless, some cities have undertaken their own surveys of rough sleepers, with varying degrees of success.104

Supported Accommodation Assistance Program (SAAP) data

 The National Data Collection Agency at the Australian Institute of Health and Welfare (AIHW) collects data on all specialist homelessness services funded by the Australian Government and states and territories.  The data contains detailed information about clients, including their reasons for requesting assistance, their circumstances immediately before and after support and the length and level of support they receive.  The agency releases the data in annual reports, as well as making an annual ‘Demand for SAAP Assistance by Homeless People’ report available and releasing regular thematic reports.

While the data captured under this collection is richer and more complex than the Census data, these services come into contact with only 19 per cent of people who are homeless on any given day.  It misses the many thousands who did not approach a SAAP service, those who use mainstream services, or people who receive a service from a homelessness service not funded under SAAP.  Also, the data collection does not provide detailed information about the medium and long-term outcomes for those who receive assistance through specialist homelessness services.

Measuring unmet demand for crisis accommodation is particularly problematic.  This information is captured by asking services to record the number of people turned away during two two-week periods each year.  While this is useful for understanding demand, it does not provide information about the circumstances of those who are turned away. 

Although there is some capacity to analyse AIHW data longitudinally, problems arise from incomplete coverage of agencies, client consent not being provided and confidentiality requirements set out in AIHW legislation which restrict access to this data.

Some agencies are reluctant to collect additional data on their clients out of concern that this will breach their privacy or because it represents an additional administrative task that prevents workers spending time with their clients. These legitimate concerns need to be balanced against the service improvements that might flow from access to richer data on clients’ pathways.

Other homelessness program data collections

Other homelessness programs administered by the Australian Government (such as Reconnect and the HOME Advice Program) or state and territory government initiatives (such as the Housing and Support Initiative and the Transitional Housing Management Program) have their own data collections.  The HOME Advice Program, for instance, provides data that includes clients’ views of their situation and goals at commencement and completion of support.  In addition, data is collected on their housing, income, employment and education circumstances before and at the end of support, on their goals and on progress made during the intervention to achieving these goals.

Information on the use of the various homelessness services or ‘pathways’ through the homelessness service system is critical to understanding how people use the service system, what interventions are successful and where improvements can be made.

Data collections in mainstream services

Mainstream agencies which are in contact with people who are homeless may not always be aware of people’s accommodation needs.  Capturing this information could inform mainstream service strategies and measure outcomes for these people.  Encouraging mainstream agencies and specialist homelessness services to adopt the same definitional approach to measuring homelessness, and to be diligent about capturing the outcomes for homeless or at risk clients, will help governments and service providers get a clear view about the true extent of homelessness in Australia. 

Data linkage across large data collections with appropriate protocols for undertaking analysis of the data would assist in understanding the complex pathways in and out of homelessness and could lead to improved responses for people experiencing homelessness.

A national homelessness research strategy

Research into homelessness is complex due to the nature of the problems and individual circumstances affecting people who are homeless.  People who may have experienced multiple disadvantages or are grappling with complex problems (such as drug and alcohol, mental health, domestic violence and poverty) may be unwilling or unable to participate in research.  Ethical dilemmas may arise for researchers about disclosing suspected child abuse or neglect or law-breaking.  The high mobility of some people who are homeless acts as a barrier to recruiting and maintaining contact with them as participants in research studies.105

The complexity of people’s circumstances and the multiple pathways in and out of homelessness can also make it difficult to examine causal relationships – randomised controlled trials or lengthy self-report questionnaires are rarely used.  Alternative methodologies such as action research or outlier studies106 to examine factors associated with resilience need to be considered. 

A further challenge in homelessness research is the state and territory differences in funding, management and service delivery models.  There is a need for multi-site or cross-jurisdictional studies, longitudinal research and comparative evaluations of practice models to determine what works best.

Homelessness research needs to engage with culturally and linguistically diverse communities, Aboriginal and Torres Strait Islander peoples and families living in rural and remote areas.  Identified population groups for research include individuals and families with substance abuse or mental health problems, people with a disability, families in which there is domestic violence and families with complex problems such as cumulative harm, dual diagnosis, multiple problems or social exclusion. 

Longitudinal studies are needed to analyse the impact of homelessness on young children, those in middle childhood and adolescence in vulnerable families.

There are significant research gaps. There is little research looking at long-term outcomes of clients assisted by specialist homelessness services. Better information is also needed on longer-term outcomes for people experiencing homelessness who use mainstream services, such as health, housing or employment services.

There is a particular need to fund research into the service system and practice issues to ensure that the services provided to people experiencing homelessness are the best they can be.  Useful areas of exploration include needs assessment and planning at the local, state and national levels, development of standardised assessment methodologies and tools, exploration of service system capacity including provider audits, unmet demand estimation and analysis of responsiveness of mainstream service systems. 

Cost benefit and cost effectiveness analyses would provide better evidence on the nature and timing of service interventions and would allow more realistic allocation of funding.  Examination of different models of multi-agency, integrated or co-ordinated service delivery to identify best practice, most efficient use of resources and outcomes is critical to service improvement.  Better quantification of the level of resources required to prevent and respond to people with complex needs who are homeless, for example, people with mental health issues or complex disability, would improve the design and delivery of services.

The Australian Government will develop a national homelessness research strategy to support the White Paper. It will build on existing research and data collection efforts.  It will strengthen the current evidence base to inform policy and practice and guide research priorities for researchers and funding bodies. 

Identifying the limitations of the existing data collections and taking steps to improve these collections will allow for more accurate measurement of the complexity of the experiences and outcomes of people who are homeless. 

As a starting point, the Australian Government will work with states and territories to pilot data linking projects in order to gain a clearer picture of the pathways through service systems for people who are homeless.

Initial projects will focus on clients most at risk, particularly children, and will explore data linkages across child protection, housing, homelessness and criminal justice systems as well as Centrelink.

Issues relating to accessibility and confidentiality of existing data also need to be addressed as part of the national homelessness research strategy so that data is more able to be used and analysed by researchers. 

State and territory governments, who have direct funding relationships with specialist homelessness services, are best placed to work directly with services to enhance data collection efforts.

As part of the national homelessness research strategy, the Australian Government will work with state and territory governments and specialist homelessness services to reform the National Data Collection process.

Priorities for reform will be:

Access to ‘real time’ data from specialist homelessness services should also be considered under the national homelessness research strategy. The availability of ‘real time’ data would avoid the need for clients to contact many agencies searching for accommodation and repeat their stories.  Access to information about bed availability in specialist homelessness services would be improved. It would also improve the match between supply and demand in specialist homelessness services.

Moving services to ‘real time’ data collection would require considerable service reform and investment in service infrastructure, particularly information technology.

The national homelessness research strategy will also focus on these additional research priorities:

The national homelessness research strategy will require collaboration from many parties, recognising that the process of reflection and engagement extends beyond the government sector.

Completed research should be publicly disseminated to inform both policy and service practice.  The Homelessness Clearinghouse recently launched by the Australian Government will be an important vehicle for this purpose; peak homelessness organisations also have an important role to play in sharing research findings.

An integrated information technology system

Connected administrative data collected from both specialist homelessness and mainstream services over time is needed. This will improve the quality and timeliness of the data collected, reduce the administrative burden on specialist homelessness services in collecting and reporting data, reduce the burden on clients in repeatedly providing the same information to different agencies and will improve both service planning and the service offer to individuals who are homeless or at risk of homelessness.   The Australian Government will undertake a scoping study to explore information technology options to better integrate homelessness specialist with each other and with mainstream services. 

Where we will start

Return to top

Chapter Seven: Implementation and Governance

Chapter Five: Breaking the Cycle