Corrective Services Family and Domestic Violence Data Project: Discussion of Findings

This report describes findings from a project undertaken by the Australian Bureau of Statistics (ABS), which explored potential options (or models) for developing national family and domestic violence (FDV)¹ statistics for Australia’s correctional services agencies. For this exploratory project, direct analysis was undertaken using administrative by-product data held in the systems of Corrective Services New South Wales (CSNSW).

One of the key purposes for this exploratory work was to identify potential data sources and methodologies that could be used to develop statistical information to support Council of Australian Governments (COAG) National Outcome Standards for Perpetrator Interventions².

The ABS was able to develop two conceptually different statistical models that could be used to derive information about FDV offenders in prison. The methodologies are presented separately, and provide a basis to determine whether either model could be practically and meaningfully developed for national statistical purposes over time.

The ABS acknowledges the considerable efforts of the corrective services agencies in each jurisdiction, including the National Corrective Services Statistics Unit Board of Management, and the Advisory Group, for their ongoing support for this project. In particular, the ABS would like to thank CSNSW for their efforts in providing experimental data and feedback regarding this project.

The ABS would also like to thank the Department of Social Services (DSS), who has funded this work under the COAG National Plan to Reduce Violence against Women and their Children 2010-2022, for their support for this project.

The ABS is interested in feedback from users of these statistics on any aspect of the release. Please send written feedback to: crime.justice@abs.gov.au.

The DSS and ABS Corrective Services FDV Data Project was undertaken in three phases. Phase 1 entailed consultation and assessment of current administrative data holdings of corrective services agencies nationally. Phase 2 involved the analysis of experimental corrective services data from the administrative data holdings of CSNSW to determine whether meaningful FDV statistics of the prisoner population could be derived. This report is the final part of this project. It includes a discussion of potential statistical models for identifying FDV offenders within the prisoner population and provides experimental results for New South Wales (NSW). A brief discussion on how to facilitate a national picture of FDV offenders in prison is also included.

The ABS recognises the difference between data collected for operational and statistical purposes. As such, the ABS acknowledges that any absence or limitation of FDV related data maintained in the current corrective services administrative systems does not represent deficiencies in the operations of these agencies, or in the data collected for administrative purposes.

Offenders serving community-based correction orders are not included in this analysis. Only prisoners held under adult corrective services authorities are considered.

When attempting to identify and measure the FDV offender population, it is necessary to understand the different methods for the identification of these types of offenders. Broadly, an FDV offender is defined in this report as a prisoner who is in custody:

• and has been charged by the criminal courts for an FDV related offence; or
• with a current active Domestic Violence Order (DVO)³
   

Please note that the definition used for FDV offenders for this project is not exhaustive and does not fully reflect all prisoners with FDV behaviours and/or tendencies. For example, prisoners with historical FDV behaviours/offending, but do not have a current FDV offence/charge or an active DVO are not in scope of this analysis.

The two models outlined in this report highlight potential opportunities in developing statistics on FDV offenders in adult corrective services custody. Both models rely on information transfers from other areas of the justice sector.

The recent releases of ABS FDV data for Recorded Crime – Victims, Australia and experimental FDV data for Recorded Crime – Offenders, Australia and Criminal Courts, Australia⁴ are based on the capacity of each agency to identify and ‘flag’ the relevant individuals within their administrative systems. Despite some variation across jurisdictions in how these flags are applied, they do provide a measure of the FDV populations within each respective justice sector (police and criminal courts).

Where these flags (or other FDV related information) can be transferred from the police and criminal court administrative systems directly to the corrections systems, a reliable measure of FDV offenders within the prisoner population could potentially be obtained.

This model contains FDV related data derived from offence level information.

In some jurisdictions, there exists specific FDV legislation for which a person can be prosecuted for FDV related matters. As such, FDV offences can be identified separately (in offence records) from other non-FDV offences.

Using offence information transferred from the criminal courts to the corrections sector, the FDV offender population within adult corrective custody consists of: Prisoners in custody with at least one FDV offence.

This model contains FDV related data derived from intervention order information.

In some jurisdictions, there exists specific FDV intervention/restraining orders which can be applied to a person for FDV related matters. This information is generally held by police agencies, and in some jurisdictions they are able to discern if a person is the victim or the perpetrator based on the type of DVO flag against their record. As such, a person can be identified as having an active DVO against them (for FDV related matters).

It should be noted that an active DVO may not always be resolved accurately in administrative systems and therefore there may be persons where their DVO remains active erroneously.

Using DVO information transferred from the police to the corrections sector, the FDV offender population within adult corrective custody consists of: Prisoners in custody with an active DVO.

Although the two models are conceptually different, both rely on information transfers from other sectors of the justice system; Model 1 from the criminal courts administrative systems and Model 2 from the police administrative systems.

Neither model will provide an encompassing dataset to answer all FDV related questions; the choice of model is dependent on the specific policy or research initiative.

For example, Model 1 could be used to understand the characteristics of persons in custody for FDV related offences; answer questions on FDV reoffending (over time); or sentence outcomes for persons with FDV related offences compared to those with non-FDV offences. Model 2 could be used to understand the offending characteristics of persons with an active DVO against them as well as evaluate effectiveness of FDV restraining orders and whether perpetrators are being held to account.

For this exploratory project, experimental analysis was able to be undertaken using existing corrective services administrative by-product data held in the systems of Corrective Services New South Wales. In NSW, specific FDV legislation exists for which a person can be charged for offences which are FDV related. This offence information is available through the NSW criminal court administrative data system. The following results have been produced for NSW using Model 1.

Approximately one in four (24%) NSW prisoners was identified as an FDV offender, meaning they were in custody as of 30 June 2016 with at least one FDV offence. (Table 1)

Overall, the experimental FDV and non-FDV populations have relatively similar characteristics across the NSW prisoner population:

• The proportion of FDV offenders who were Aboriginal and Torres Strait Islander was 32%, compared to 22% for non-FDV offenders.
• The proportion of FDV offenders who were Australian born was 85%, compared to non-FDV offenders who were born in Australia which was 74%.
• FDV offenders were more likely to have been previously imprisoned (56%) than non-FDV offenders (50%). For NSW prisoners overall, 52% had previously been imprisoned. (Table 2)
• FDV offenders were younger than non-FDV offenders, with an average age for 35.2 years compared to 36.9 years. The average age for all NSW prisoners was 36.5 years. (Table 3)
   

The ABS continues to develop a range of FDV statistics for the justice sector. The ABS has presented two statistical models which can be used to identify FDV offenders within a jurisdiction’s prisoner population. Model 1 has been applied to CSNSW administrative data to produce experimental FDV statistics.

Whilst each model has its own caveats and limitations, they provide an insight and a starting point to developing a national evidence base for FDV offenders within the prisoner population. In further advancing this evidence base, the ABS will continue to work in partnership with corrective services agencies in each jurisdiction.

This release previously used catalogue number 4540.0.00.001