4529.0 - Conceptual Framework for Family and Domestic Violence. , 2009  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 18/05/2009  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All  



Prevalence and incidence

Levels of victimisation in society through family and domestic violence can be measured in multiple ways, and incidence and prevalence are both terms used in this Framework as measures of the extent to which members of the community may be victimised by FDV:

  • prevalence: defined as the number of people in the relevant population who have experienced FDV at least once in a given period. Prevalence estimates are measures of the extent of victimisation within the community;
  • incidence: the number of incidents of FDV that have occurred in the relevant population in a specified reference period. This measures the extent of offending behaviour within a community. Because many FDV victims experience more than one episode, incidence measures will be larger than prevalence measures.

These terms can be used in a range of different ways, depending on the FDV definition that is most relevant, and the context in which the measure is to be used.

Data limitations

Family and domestic violence often takes place in the private domain, and can comprise a range of illegal and legal actions and events. Any attempt to collect reliable information will encounter problems arising from issues of perception (whether an incident was FDV-related and thus whether, under a legal FDV definition, it involved the commission of one or more acts against another person or persons), and therefore of self-classification by both the victim and the perpetrator. Other issues that limit attempts to measure items of interest about FDV include the following:
  • under-reporting: many crimes are not reported to police or other authorities, so neither the total number of victims nor the total number of perpetrators is captured in their data;
  • hidden reporting: this may occur where a victim seeks services, or reports an incident, but does not disclose that FDV was the reason for the contact;
  • under-recording: this occurs due to variations in the processes followed for recording incidents coming to the attention of authorities or services. There is also potential for misclassification of FDV incidents, e.g. where a victim presents as a general assault victim (not an FDV victim); and
  • counting or recording rules: as an example, some jurisdictions’ FDV-related orders also cover non FDV-related disputes, thus overstating the number of orders handed down for FDV perpetrators. For example, in Victoria, an Intervention Order may be handed down for FDV and non FDV-related incidents, such as stalking. Conversely, Apprehended Violence Orders in NSW are linked to FDV incidents. As a result, Victoria's count of Intervention Orders may not provide a good indication of FDV orders and overstate the use of these for FDV incidents. In the health system, treatment for specific injuries may be recorded without recording the cause of the injury.

It must be acknowledged that it is unlikely that the ‘real number’ of incidents of FDV will ever be known. Different collection vehicles, using different scopes and definitions, will often produce different levels of estimates. At the same time, these methodological issues that apply to counts of criminal incidents are compounded by the complexities of constantly evolving definitions of FDV. However, data that are currently available may provide valuable estimates that can be used as indicators in the development of a picture of FDV within the context of the Conceptual Framework.

Further, it is important that when measuring the growth or reduction in FDV or in assessing the impact of different interactions that a common definition and collection method is used. This can be particularly difficult when the legal and other definitions used by authorities and agencies are changing over time.

Data sources

Any attempt to build a picture of FDV in Australia must look to include information from a number of sources. These sources can vary in quality. It is also recognised that there are many potentially useful data sources that are relatively under-utilised or are not in the public domain. Work that could be undertaken using this Framework could include the assembly of an inventory of these data sources, their definitions, methodology and quality.

The two main types of data sources that can be used to obtain information about FDV are administrative by-product data and surveys.

Administrative by-product (ABP) data can be derived from government agencies (e.g. Police, Family Courts, Child Protection Authorities) and service providers (e.g. Family Relationship Services Programs) that keep clinical, case or other administrative records about clients and the nature of their transactions with the service. Information may be extracted from these records and compiled for statistical or analytical purposes. There are advantages in using ABP data, as direct respondent burden is minimised and data are often able to be collated or provided to a central data collection agency electronically. The usefulness of ABP data can be limited by the fact that it cannot account for people who do not report to or access the services provided, and because the data focus on attributes relating only to that specific service. Additionally, the primary role of these agencies is that of service providers - not as repositories or collectors/providers of data. The nature of their services is therefore such that not all elements of the FDV Framework could be covered through the use of administrative data. This is further complicated when authorities and agencies do not record the separate elements in the incident, thus preventing comparison of data from different sources if different definitions of FDV are used.

One of the primary reasons for conducting victimisation surveys is to collect data not captured by ABP, as many victims of crime do not report their experiences to the police or other service providers. As surveys use a standard definition of incidents, they are able to measure FDV incidents that may or may not be classified as criminal across all jurisdictions or by all service providers. Comparability issues exist when attempting to align rates of FDV victimisation. There are variations in legislation between jurisdictions and services, and administrative data therefore relies on differing definitions and thresholds. Surveys are therefore useful in their ability to overcome this data issue. However surveys are still limited in that they are subject to problems of recall, disclosure and coverage. For example when being surveyed, respondents may not remember an incident of victimisation or use of a relevant service that could be reported and they may not be comfortable speaking about an incident that was particularly upsetting or confronting. No single data source can ever provide a complete picture of the occurrence of crime or the uptake and usage of services.

Previous PageNext Page