Chapter 5

Chapter 5

Improving the evidence base

5.1        In its 2009 report, Time for Action, The National Council to Reduce Violence against Women and their Children (National Council), described the data relating to violence against women and their children in Australia as 'poor'.[1] Throughout the inquiry the committee received evidence emphasising the lack of data available on the prevalence and impact of domestic and family violence. For example, the Women's Centre for Health Matters stated:

Despite the existence of large-scale data collection mechanisms like the [Personal Safety Survey] and [Australian Institute of Criminology], it's evident that there are still major gaps in our understanding about the prevalence and impacts of domestic and family violence.[2]

5.2        As was discussed in Chapter 4, the Australian Bureau of Statistics' (ABS) Personal Safety Survey (PSS) provides national data on domestic violence, however it was criticised for its failure to adequately sample from subgroups within the population, such as women with disabilities, women from culturally and linguistically diverse backgrounds, immigrant and refugee women and Indigenous communities.[3]

5.3        The National Council highlighted the need for robust data collection systems to support prevention and early intervention services.[4]

5.4        In the course of this inquiry, Women's Health Victoria noted the importance of continued collection and analysis of data about the impact of family violence:

It is important that data about the impact of domestic violence...continue to be collected and analysed so that we can maintain an accurate picture of its prevalence, and its health and social impact.


Effective data collection can illustrate whether the systems are meeting the needs of victims and further identify and highlight gaps in policy and services.[5]

5.5        Similarly, the Aboriginal Family Law Services (WA) emphasised the need for the 'development of a more coherent data collection system and evidence base':

From a foundation of reliable and consistent data, we will gain a more accurate picture of how broadly this issue impacts on communities, particularly for Aboriginal and Torres Strait Islander people, and the remedies we can employ to eliminate and prevent violence.[6]

Barriers to data collection

5.6        The National Council summarised why collecting and analysing data on domestic and family violence has been difficult:

Data on services sought by, and provided to, victims is not readily available, and the way in which information is reported is generally inconsistent and does not allow for a comprehensive understanding of violence against women. Variations in data estimates across Australia are affected by differences in what is captured, counted and reported across States and Territories.

There are also personal and institutional barriers in decision making within and across systems that reduce the extent to which sexual assault and domestic and family violence is disclosed and reported. This affects the capacity of data to accurately reflect the real numbers of women and children who experience this violence. The difficulty in measuring the true extent of sexual assault and domestic and family violence in the community has been widely acknowledged.[7]

5.7        These issues are discussed further below under two broad categories, namely:

Reporting domestic and family violence

5.8        Evidence to the committee highlighted that the data available on the prevalence of domestic and family violence is generally an underestimate because many occurrences go unreported. For example, Professor Donna Chung, in a paper for White Ribbon Australia – Understanding the Statistics about Male Violence Against Women, emphasised that all statistics will underestimate the actual extent of the problem:

At the outset, it is important to note that all statistics about [male violence against women], regardless of their source, will be a conservative or under-estimate of the actual extent of the problem. This is because there will always be women who are understandably distressed or embarrassed about having been subjected to such violence, and as such, do not disclose or report it.[8]

5.9        Mr John Paterson, Chief Executive Officer, Aboriginal Medical Services Alliance Northern Territory, also noted the true extent of violence is under-reported:

Aboriginal people may not report violent incidents if doing so will result in a family member being removed from the community or incarcerated. Women may also not be willing to report violence out of fear of having their children removed from their care by child protection authorities.[9]

5.10      The evidence to the committee indicates that it is difficult to assess the extent of this underreporting. For example, the Australian Women Against Violence Alliance (AWAVA) cited research from 2011 that estimated 90 per cent of cases of domestic violence in Australia went unreported. Furthermore, AWAVA noted:

A 2005 report found that in the twelve months preceding the research period only 5% of women who had experienced violence from a current partner had reported the last incident to police. This demonstrates that current domestic violence statistics are a conservative estimation of the prevalence of intimate partner violence and that actual rates of violence are estimated to be much higher.[10]

5.11      The Redfern Legal Centre referred to 2012 research suggesting that only 50 per cent of victims of domestic violence report the abuse to the police.[11]

5.12      The ABS observed that rates of reporting domestic and family violence have improved over the last decade; however, estimates still suggest many incidents still go unreported.[12] The ABS outlined some of the barriers that may prevent a victim from disclosing an incident of domestic and family violence and seeking help:

5.13      A few submissions also referred to 'hidden reporting', where a victim seeks assistance from a service but does not disclose domestic and family violence as the reason for making contact with a service provider.[14]

Uniformity and consistency of data collected

5.14      The National Plan explains why the problem of a lack of uniformity and consistency in data occurs:

Jurisdictions collect and report different administrative data on experience and perpetration of violence against women and their children. This data is collected through systems such as policing, justice, corrections, health and community services. These systems are often not 'linked-up', meaning the individual pathways of women and their children experiencing violence, and of perpetrators, cannot be tracked across systems. This presents a considerable barrier in determining which interventions are most effective in supporting and protecting women.

Data is also often not comparable across jurisdictions, due to different data definitions and collections. Making data consistent, and developing a national picture around administrative data, is important in better understanding the incidence and experience of violence against women and improving interventions.[15]

5.15      Mr John Hinchey, the ACT Victims of Crime Commissioner, described this problem as 'we get what we can rather than define what we want' in terms of data collection:

Because there is no uniformity around data collection we are reliant on each individual agency's data collection capacity. Therefore, we seem to be approaching things from the wrong end. We get what we can rather than define what we want. We are uncertain what this is telling us...We are at a little bit of a loss at times as to what it is we are actually wanting to find out, and how to find that out.[16]

The National Plan

5.16      The National Council explained how, in its view, the deficiency in data collection needs to be addressed through the National Plan:

A national minimum data set needs to be developed (including a data dictionary and standard protocols) to enable consistent and standardised data collection methods and analysis for sexual assault and domestic and family violence. The development of common on-line databases that have the ability to monitor individuals across the service sector and across jurisdictions (with, for example, the use of a unique identifier) will also support accurate and meaningful national data collection.[17]

5.17      The National Council also proposed the establishment of a 'National Centre of Excellence for the Prevention of Violence against Women':

A centralised, independent, and expert capability is needed to coordinate evidence building and sharing through research, data collection, data analysis, monitoring, evaluation and review...This body would:

5.18      To this end, one of the 'foundations for change' in each of the Action Plans making up the National Plan will improve the evidence base.[19] This will be done through:

5.19      The National Data Collection and Reporting Framework and the National Centre of Excellence – now known as Australia's National Research Organisation for Women's Safety (ANROWS) – are discussed below.

National Data Collection and Reporting Framework

5.20      The National Plan states:

[T]he evidence base for work in domestic violence and sexual assault will be improved through all jurisdictions' commitment to a national data collection and reporting framework. In the long term, the aim is to create nationally consistent data definitions and collection methods.[21]

5.21      The data framework will be operational by 2022.[22]

5.22      Under the First Action Plan, the ABS has worked with governments in the early stages of developing the National Data Collection and Reporting Framework. The ABS has also released two documents looking into certain aspects of data collection:

5.23      However, the Implementation Plan for the First National Plan cautions 'it is unrealistic to expect consistent data within the first three-to-four years of the National Plan'.[24]

5.24      The Second Action Plan continues the work to develop the National Data and Collection Reporting Framework:

The framework will work with existing national data collected from state and territory systems and lay the foundation for building a common language and a coordinated and consolidated approach to data collection.[25]

5.25      It is envisaged that over the period of the Third Action Plan, governments will have use of the improved data:

The Third Action Plan will deliver solid and continuing progress in best practice policies, with governments using data of far greater detail, accuracy and depth due to the improvements made in data collection and analysis.[26]

5.26      The Department of Social Services stated in its submission:

The Commonwealth Government has allocated more than $100 million over the next four years to support the Second Action Plan...[and] around $200 million has been committed to address violence against women and their children between 1 July 2009 and 30 June 2017.

This funding includes the following...$1.7 million to take the next steps in developing a national data collection and reporting framework, including $300,000 for the Australian Bureau of Statistics to augment data sets on victims and offenders.[27]

Support for the National Data Collection and Reporting Framework

5.27      There was support for the development of a National Data Collection and Reporting Framework. For example, Destroy the Joint stated:

The proposed National Data Collection and Reporting Framework and related research efforts from the Australian National Research Organisation for Women's Safety (ANROWS) is critical not only to ensure the safety of women and children and others impacted by domestic violence in the community, but also to communicating the true prevalence and impact of domestic violence in Australia. In relation to the issue of reporting, recording and monitoring data, we specifically commend Priority 5 in The Plan [continuing to build the evidence base].[28]

Challenges to establishing a National Data Collection and Reporting Framework

5.28      Although there was support for the National Data Collection and Reporting Framework, a number of challenges to its establishment were identified. For example, the South Australian Government noted that the National Data Collection and Reporting Framework will require commitment at all levels of government.[29]

5.29      The committee also received evidence expressing concern about the resourcing of data collection. For example, Ms Marcia Williams, Chair of the ACT Domestic Violence Prevention Council, referred to the importance of the data collection and also to the difficulty it poses in terms of resources:

For us, the evidence collection in the second plan is really critical. We do not have the capacity to get a lot of data. It is a hard thing to do at a local, ACT level, even though we are small and we should be able to do it. For us it is making sure that the national approach actually supports all of the local jurisdictions as well, so we can get that data to understand the real situation and influence it.[30]

5.30      Mr John Hinchey, the ACT Victims of Crime Commission, expressed concern that a lack of resources generally hampers efforts to coordinate data collection:

I think the agencies are currently collecting data. They are not going to be able to come together and reach agreement around benchmark data collection and data sets and then move forward on that, because they are not resourced to do it. No-one is resourced to pull this together to coordinate it, to do a literature search, to work with the ABS, to come up with a framework of data collection and then to monitor it, ensure that the data is collected and then reported on quickly. I do not like harping about the lack of resources, because it is a disempowering position to be in and it prevents people from getting things done.[31]

Improving data collection

5.31      The committee received suggestions for potential improvements to the categories of data to be collected. For example, Our Watch suggested that the data collection methodologies could be adapted for use at a local level:

Local governments have expressed interest in undertaking surveys to establish their own baseline for prevention of violence against women and their children and gender equality strategies. A system and tools to make survey questions, and support for their delivery and analysis, available to local governments and regions would provide enormous benefits to being able to demonstrate progress in prevention as a result of local action.[32]

5.32      The committee received a number of submissions calling for the disaggregation of data, particularly in respect to vulnerable groups.[33] For example, the Federation of Ethnic Communities' Councils of Australia argued:

Disaggregated data and research on the prevalence of domestic and family violence within culturally and linguistically diverse communities should be collated at a national level to determine the rates of violence and the different variables and factors that influence it, including cultural or ethnic background, economic status, level of education, religious/cultural beliefs and location. The collected data and research should subsequently inform the development of a targeted national strategy to tackle violence across diverse communities.[34]

5.33      Similarly, the Inner City Legal Centre contended:

Data collection and reporting, disaggregated for gender and LGBTIQ status, should be a priority for the judicial system and service providers at a state and national level to enable research to be undertaken on the prevalence and impact of domestic violence in LGBTIQ communities.[35]

5.34      Women's Centre for Health Matters expressed the view:

Until these mechanisms [for large-scale data collection, such as the PSS] are complemented by more detailed, cross-tabulated data from sources that capture vulnerable, isolated population groups and people who unlikely to report violence to authorities—data that is able to be disaggregated by data items such as locality, disability status, gender identity, and so on—then our understanding of the prevalence and impact of domestic and family violence remains limited.[36]

5.35      The ACT Women's Services Network called on the Commonwealth Government to ensure that the ABS was adequately funded and resourced:

[T]o provide the gender-disaggregated and cross-tabulated State/Territory data that is necessary for us to have meaningful data and to establish the rates of violence including against vulnerable groups like women with disabilities, women from culturally and linguistically diverse backgrounds, Aboriginal and Torres Strait Islander women, and gay, lesbian, transgender and intersex people.[37]

5.36      Destroy the Joint called for 'an official information page [to] be established where Australians can access accurate, unbiased data and facts on domestic and family violence in a format which is easily understood'.[38]

Australia's National Research Organisation for Women's Safety

5.37      The National Plan states that '[a]ll governments recognise that outcomes for women and their children could be improved by governments working more collaboratively through building the evidence base, sharing information and tracking performance'.[39] The role and function of ANROWS, as articulated in the National Plan, is to:

[B]ring together existing research, as well as undertake new research under an agreed national research agenda. Through pursuing research in a cohesive national way there will be greater opportunities to support research which is more responsive to policy makers' and service providers' needs. National research will fill gaps in knowledge and help increase the understanding of issues across different sectors such as health, justice, education and housing.[40]

5.38      ANROWS was officially launched in May 2014 as an independent, not-for-profit organisation, jointly funded by the Commonwealth and all state and territory governments.[41]

5.39      In its submission ANROWS described its mission as:

[T]o deliver relevant and translatable research evidence which drives policy and practice leading to a reduction in the incidence and impacts of violence against women and their children by 2022.[42]

National Research Agenda

5.40      One of the priorities of the Second Action Plan is continuing to build the evidence base, including expanding the quality and quantity of national research on violence against women and their children through the implementation of the National Research Agenda:

In the second half of 2013, ANROWS conducted considerable consultation across sectors to inform the development of the National Research Agenda to shape and guide national research on violence against women and their children.

The National Research Agenda was endorsed by all Australian governments and released on 16 May 2014. It will inform the development of research by a range of institutions, academics, governments and community organisations.[43]

5.41      The National Research Agenda is organised into four 'Strategic Research Themes':

  1. experience and impacts;
  2. gender inequality and primary prevention;
  3. service responses and interventions;
  4. systems.[44]

5.42      The Second Action Plan states:

Common across all Strategic Research Themes is the need to focus research effort on "what works" and on diverse groups and under-researched populations, including Indigenous women, women from CALD communities and women with disability.[45]

5.43      At the public hearing in Sydney, Dr Mayet Costello, Research Manager, ANROWS, provided the committee with the following information on work ANROWS is undertaking to support the National Research Agenda:

[On 31 October 2014 ANROWS] launched our first-ever research program, which is for the 2015-2016 financial year...We have 20 projects that we launched [with] a combined total value of approximately $3.5 million, so it is a very large and ambitious research program. It is probably the biggest in this area in Australia. We have a really ambitious reach with our research program—we have sites in every Australian state and territory, including a number of national projects with sites across the country. We have a spread focusing on both types of violence—sexual assault and domestic and family violence—as well as particular priority population groups that have been identified under the national plan and the national research agenda. We have particular projects on Aboriginal and Torres Strait Islander women, women with disability, women from culturally and linguistically diverse backgrounds and women from rural and remote areas; as well as projects on other issues like younger women and the correlation between mental health and drugs and alcohol and violence against women. We are currently in the process of establishing a perpetrator interventions research stream with dedicated funding from the Commonwealth government, which is $1 million per annum over a three year program.[46]

5.44      Dr Costello also noted ANROWS has a number of potential further research projects which are currently unfunded:

We have a waiting list of eight projects, which are very worthy and very interesting projects, and we are hoping that if funding is identified throughout the financial year—through savings and other measures within our organisation—we will be able to fund additional projects. We are planning to release the second stage of our research program in about the middle of next year with whatever else we can put together. The remaining eight projects have an approximate value of $1.8 million.[47]

Funding for ANROWS

5.45      In its submission, ANROWS noted that it is only funded for three years, receiving $3 million per annum for the period 1 July 2013 to 30 June 2016, which means its funding is due to expire six years before the end of the National Plan.[48]

5.46      Dr Costello explained to the committee that the short period of funding presented some challenges to ANROWS' research work:

The short funding period is a little bit of a challenge for ANROWS in that building a rigorous, robust and academically credible research program is a bit of a challenge in the shorter term, particularly given that longer-term research, such as longitudinal research, is really important for understanding the effectiveness of programs and service delivery, such as programs working with men who use violence. Unless we do research before, after and some time after an intervention it is very hard to actually provide compelling evidence on effectiveness.[49]

5.47      Dr Costello identified two key areas which would benefit from longitudinal research, namely prevention research and service intervention:

In terms of prevention...the emerging research on prevention tends to look at process oriented evaluation. Was a program run well? Did people fill out their evaluation forms? Did people engage? Did people attend? What is not as well understood and cannot be followed up without longitudinal research is if those initiatives or programs result in behaviour change and/or attitude change, and is that sustained over time? Some of the very preliminary research suggests there are mixed evaluation results at six months and/or two years post intervention. It is absolutely crucial. Process will only tell you so much and it will only tell you whether or not something was well run. It will not actually tell you if something was effective. If we want to make a change to this issue—if we want attitudinal change and we want behavioural change—then we need to follow up at longer periods of time post intervention.

Similarly, with men who use violence—and bearing in mind that our perpetrator interventions research is very much in its nascent stages—there is the similar issue, particularly for interventions that are court mandated or socially mandated...What we know is that the closer they are to that mandate, such as court order, the more likely they are to comply. What we do not know is post that mandate—and even during that for some men—how effective the intervention is. How likely are they to repeat or to reoffend in terms of their violence? So longitudinal research that looks at effectiveness and outcomes is important. One of the key things for both prevention and interventions with people who use violence is that we do not have a good quantum of effective measures for what constitutes success.[50]

5.48      Dr Costello indicated that some of ANROWS' projects have research timeframes beyond ANROWS June 2016 funding:

Again, recognising the need for a very strong research program, we have determined that two years is the maximum that we can support. That means that, technically, a number of our two-year programs or projects will extend beyond the date of our funding agreement. We made it until the end of December 2016 with the understanding that if we were not funded past that point then we would be able to transfer; our constitution has provisions for closure, and we would be able to transfer some of those contracts to a like organisation and/or to the Commonwealth government potentially to finish those contracts. So it is an issue.[51]

5.49      ANROWS' submission argued for a longer-term funding commitment:

A longer-term funding commitment, at least to the end of the National Plan in 2022, is necessary to enable ANROWS to fulfil its potential including support for longer term research projects, which are crucial in understanding, for example, the effects of perpetrator intervention programs. To illustrate this point, the recent open grants applications process conducted by ANROWS, resulted in 50 applications for research projects to address current gaps in the evidence base with a total value of approximately $15 million.[52]

Support for ANROWS

5.50      A number of submissions supported the establishment of ANROWS within the framework of the National Plan. For example, the Central Australian Women's Legal Service stated:

We welcome the establishment of national responses to domestic violence such as the National Plan and its associated Action Plans, as well as the related initiatives including the establishment of the Foundation to Prevent Violence Against Women [Our Watch] and the Australian National Research Organisation for Women’s Safety (ANROWS).[53]

5.51      The submission on behalf of the Tasmanian Government stated:

It is important that the National Plan continues to support its flagship activities including the Foundation to Prevent Violence Against Women and their Children [Our Watch] and [ANROWS] to build the evidence and best practice in primary prevention that will support the needs of women living with disability, Aboriginal and Torres Strait Islander women.[54]

5.52      While there was support for ANROWS, there were also reservations expressed about its work. Associate Professor Dea Delaney-Thiele, Chief Executive Officer of the National Aboriginal and Torres Strait Islander Women's Alliance (NATSIWA), argued that NATSIWA should be involved in ANROWS' work on Aboriginal and Torres Strait Islander communities:

If not [NATSIWA], it needs to involve the communities. The community needs to be part of a process...I only make the comments about it because I believe that research needs to be separate from government and be an independent voice.


I really believe that there needs to be Aboriginal governance and control over the health research agenda.[55]

5.53      The Multicultural Centre for Women's Health cautioned that ANROWS' work was only part of the necessary research required:

Comprehensive and detailed research is needed so that violence prevention initiatives may be evidence based. There has...never been any comprehensive nationally-focused research that investigates the specific experiences of violence of immigrant and refugee women as a group in Australia. While [ANROWS] recently called for submissions from researchers and community groups to conduct research into the incidence and prevalence of violence against immigrant and refugee women, the findings of only one research project will not provide the evidence base needed to inform response, early intervention and prevention programs and strategies across Australia. Further and more diverse research, providing both qualitative and quantitative data, and exploring the full range of issues, across the full diversity of women as they vary according to geography, culture, migration status, age, ethnicity and socio-economic status.[56]

5.54      The committee also received recommendations for specific areas on which ANROWS should focus its research. For example, Women with Disabilities Victoria commented:

That the Australian Government and ANROWS [should] support research into people who choose to use violence against women with disabilities across the range of domestic settings they live in, in particular with regard to residential care settings. Research can inform practice guidelines for services, violence responses and preventions.[57]

5.55      Women's Centre for Health Matters identified two areas for further research by ANROWS:

ANROWS [should]:

undertake research to create a national definition of gender equality, a vision of what success would look like, and strategies for achieving it; and

undertake research that enhances our understanding of which aspects of gender inequality have the greatest impact on the prevalence of domestic violence.[58]

Committee view

5.56      The committee strongly agrees with witnesses as to the importance of effective national data collection and research in order to determine appropriate policies to address domestic and family violence. The committee notes the limitations on data regarding the prevalence and impact of domestic and family violence, and supports the measures in the National Plan to improve the evidence base.

National Data Collection and Recording Framework

5.57      The committee understands that the development of the National Data Collection and Recording Framework is in its preliminary stages. The committee appreciates that under the National Plan jurisdictions have agreed to have the framework fully operational by 2022 and notes that, realistically, consistent data under the framework is unlikely to start to be generated and used until the period of the Third Action Plan, that is 2016-2019.

5.58      The committee is supportive of the initiative to collect nationally consistent data, however, the committee shares the concerns of witnesses that a lack of resources could, potentially, be a constraint on agencies ability to collect and collate data pursuant to the framework.

Recommendation 5

5.59      The committee recommends that the Commonwealth Government lead and coordinate the work to facilitate data collection pursuant to the National Data Collection and Reporting Framework.

5.60      In the committee's view, Our Watch's suggestion that a system and tools for making survey questions, delivery and data analysis available to organisations such as local governments appears reasonable. Given the strong emphasis on developing a consistent data collection framework in the National Plan, and the work already carried out by the ABS, it seems sensible to enable organisations to undertake the collection of data on domestic and family violence where they are willing and have the resources to do so.

Recommendation 6

5.61      The committee recommends that the Australian Bureau of Statistics, and other relevant organisations, investigate the feasibility of developing systems and tools which would enable survey questions, delivery and data analysis developed pursuant to the National Data Collection and Reporting Framework to be modified and made available for organisations to use on a local level.

5.62      The committee received a number of submissions calling for the disaggregation of data on domestic and family violence according to specific categories of information. Given that the National Plan envisages governments 'using data of greater detail, accuracy and depth'[59] by the period of the Third Action Plan, the committee assumes that the work currently being done by the ABS to develop the National Data Collection Reporting Framework would ensure that the data identified for collection is able to be disaggregated in a wide variety ways. However, for the record, the committee encourages the ABS to work with interested stakeholders to address their needs in terms of the disaggregation of data.

Australia's National Research Organisation for Women's Safety

5.63      In the committee's opinion, the establishment of ANROWS is a key initiative under the National Plan. The committee believes that the results and findings from ANROWS' research program will make a significant contribution to 'fill[ing] gaps in knowledge and help increase the understanding of issues'.[60] The committee is pleased to note the ANROWS research program which includes projects on Aboriginal and Torres Strait Islander women, women with disability, women from culturally and linguistically diverse backgrounds and women from rural and remote areas.

5.64      However, in the committee's view, the fact that ANROWS only has funding until June 2016 is disappointing. Especially considering that this timeframe for funding means that some projects in ANROWS' current (and first) research round do not even have funding certainty for the entirety of a two-year project. The committee also believes that there is a strong case for funding longitudinal research into prevention and intervention initiatives.

Recommendation 7

5.65      The committee recommends that the Commonwealth Government provide necessary secure funding to ANROWS until at least the end of the implementation of the National Plan in 2022 to provide for the continuation of its research work and to enable ANROWS to conduct longitudinal research.

Navigation: Previous Page | Contents | Next Page