While Victoria Police has made considerable changes in updating its policies, guidance and training to reflect the need to accurately identify the predominant aggressor, this has not consistently translated to improved police practice.
Internal Victoria Police analysis of a sample of incidents in 2020 where a woman was named as the respondent suggests that it misidentifies female respondents at a rate of 12 per cent. A further 35.2 per cent of cases can be classed as ‘borderline’ for correct identification. Professor Cathy Humphreys advised us that in previous research analysing police Family Violence Reports where a woman was a respondent, almost half (48 per cent) of cases involved misidentification.27 28 There is no doubt that officers face extremely challenging situations when they attend family violence incidents, and the predominant aggressor will not always be plainly obvious. However, there are some clear process and practice issues that need further attention. We must acknowledge that, based on substantial internal work Victoria Police has done to understand the issue of misidentification, they themselves have identified many of the issues raised with us by stakeholders.
Police must nominate the perpetrator at the incident, even if they are unsure
At the beginning of the police risk assessment, the Family Violence Report, which police must complete for every family violence incident they attend, they must nominate the respondent (perpetrator, predominant aggressor) and affected family member (person most in need of protection, or victim).
Victoria Police advised us that officers will generally speak to both parties to reach a view about who is the victim and who is the perpetrator. Observation, physical evidence and witness statements also inform their decision. However, there are cases where police are genuinely unsure, whether as a result of family violence knowledge gaps, or the complexity of the case. In such a case, they make a best assessment (refer to Tracey’s story). This is particularly true for LGBTIQ+ family violence, which stakeholders told us was more likely to be mistakenly seen as mutual, as a result of poorer understanding of the dynamics of violence, coercion and control in these relationships on the part of police.
… police are more likely to fail to identify a predominant aggressor in situationally ambiguous circumstances, for example when both parties tell police they are the victim of the other. Additionally, case study data indicates that victims are more likely to be misidentified when police work demands are privileged over investigatory ones, or when trivial criminal information, such as a victim’s use of an illicit substance, supersedes [intimate partner violence] risk-relevant information.
No to Violence (2019): Discussion Paper: Predominant Aggressor Identification and Victim Misidentification, p. 6
Furthermore, Victoria Police estimates that in about a third of cases, officers only speak to the person they believe to be the affected family member. This can mean they accept the story presented by a perpetrator claiming to be a victim, and the actual victim survivor misses an important opportunity to challenge this account.
These uncertain nominations of respondent and affected family member have real consequences because once a sergeant signs off on the Family Violence Report and it is committed to the LEAP database, it remains on record permanently. Sound internal review processes at Victoria Police are therefore vitally important to pick up on misidentification.
Furthermore, the Family Violence Report and/or its guidance may need additional prompts to more accurately identify the predominant aggressor and person most in need of protection, particularly in ambiguous situations. The continuing release of the MARAM perpetrator tools and guidance provides an opportunity for such a review in line with agreed Victorian risk assessment and management processes [relates to action 4].
Tracey and Celina were in a relationship. There was a history of Celina using violence in the relationship, but Tracey had not accessed or been referred to family violence services or police before.
Police were called by passers-by to an incident where they found Tracey and Celina outside their house, in a heated argument, punching each other. The police members attended, separated the two parties and spoke to them separately. They did not explore the context of previous abuse in the relationship.
No criminal charges were laid, but police issued a family violence safety notice at the scene, listing Tracey as the respondent and Celina as the affected family member. They noted that they were unable to determine the predominant aggressor in this instance.
At the first hearing in court, the police and magistrate agreed that the FVIO application would be withdrawn, effectively dismissing the seriousness of the violence. Knowing this, Tracey decided not to say anything about the abuse she had been experiencing in the relationship. Due to this withdrawal, no further risk assessments around the relationship were completed. As a result, an opportunity was missed for the women to be referred to relevant family violence services. If police are ever called to an incident again where Tracey is present, police records will show that she was a perpetrator in a past incident.
Source: Family Violence Reform Implementation Monitor, based on a case example provided by a community legal centre
Police do not always sufficiently consider the history and circumstances of an incident
All stakeholders consistently raised concerns that, too often, police are taking what they see and hear at the incident at face value and are not digging deeper to understand the context in which an incident occurred.
Stakeholders advised that this appears to be the result of varying levels of understanding of family violence, biases and stereotypes among some officers, and inconsistency in adherence to police policies and processes. This results in them not understanding the factors that can contribute to misidentification (such as manipulative perpetrators and the characteristics of victim survivors – as outlined in section 4). These issues have been highlighted in recent news articles,29 and while some of these stories are from other Australian states, they echo the Victorian stories we heard during our consultations. We heard numerous examples from community legal centres and directly from victim survivors. The key factors in many of the misidentification cases were:
- Police listened to the calm and well-spoken male perpetrator, believing him to be the victim, including in cases where the woman was the one to call police and request help but was hiding or had left the house once police arrived. In these cases, women often didn’t know they’d been named as the respondent in the incident and were very distressed to find out when they were served with an FVIO application.
- Police seemed to overlook signs of trauma and self-defence among those in need of protection.
- Police also seemed to either not realise or not give sufficient regard to the fact that the woman they were listing as the respondent often had no prior offending and often had a documented history of abuse by her partner (the alleged victim) or another partner.
As well as these general issues, there were further issues for minority groups that affected their ability and/or willingness to disclose family violence, which seemed to play a large role in police misidentifying them:
- For non–English speaking and migrant women, police seemed to lack an awareness of their fear of speaking up due to the perceived risk of being deported. Cultural differences also possibly contributed to a reluctance to speak openly about family violence, including particularly sensitive topics such as rape. Very careful questioning is required to uncover what is happening in these situations. In some cases, though, police used the perpetrator to interpret for a non–English speaking victim, which goes against police policy.
- Aboriginal women are often deeply mistrusting of police due to past experiences of systemic racism and the perceived risk of Child Protection becoming involved and children being removed from their care. They are also statistically more likely to have a traumatic brain injury, with Aboriginal women experiencing head injury due to assault at a rate 69 times the rate for non-Aboriginal women.30 All of these factors will affect their communication and cooperation with police, yet stakeholders told us that this continues to be insufficiently acknowledged by some police. In some cases, police interviewed Aboriginal women in front of the perpetrator and dismissed violence against Aboriginal women in smaller communities because they believed the couple would ‘be back together tomorrow’.
- There was a clear view that police in regional and rural areas had a poorer understanding of the dynamics of family violence and that it is particularly difficult to get an appropriate response in these areas if the family is known to police.
- There was a perception by stakeholders that police don’t take certain voices seriously, particularly those of Aboriginal women and children and young people, treating them as unreliable.
Family violence is core business for police: 40 to 60 per cent of Victoria Police callouts on any given day are to family violence incidents.31 In 2021, 47 per cent of all crimes against the person related to family violence.32
Despite the enormous effort of Victoria Police through existing family violence training offered through the Victoria Police Family Violence Centre of Learning, there appears to be a real need for further targeted capability building activities for frontline police and sergeants to ensure they truly understand the complexity of family violence and how it might be experienced by a range of marginalised groups [relates to action 14]. Better understanding these dynamics could help police to work more effectively with vulnerable groups to:
- overcome their hesitancy to disclose family violence
- be more alert to perpetrators’ manipulative tactics
- have the skills to engage with likely perpetrators in a non-collusive way.33
There is likely to be value in involving the specialist family violence sector, which has not previously been involved, in the future development and delivery of police training, targeting known issues. This should include Aboriginal family violence organisations [relates to action 13].
Victim survivors who use force are often labelled perpetrators
Despite police guidance being clear that a victim who uses self-defence or violent resistance against a perpetrator is not themselves a perpetrator, there appears to be a real practice tension for police in situations where a victim uses force. This tension arises, in part, from the dual functions of the Family Violence Report that is completed about the incident.
Susan and Alex are married and were living together with older children from previous relationships and a baby they had together. Alex worked full-time and Susan was studying law and working in a part-time role in the criminal justice system. Neither had a criminal history nor any previous dealings with the police. Alex had a pattern and history of exercising coercive control over Susan and was occasionally violent. Alex’s main ongoing abuse was manipulation through threats and denying Susan access to money and assets of the relationship.
On the night of Susan’s arrest, they had both been drinking and arguing about money from the proceeds of a house sale that Alex had attempted to hide in his mother’s name to obstruct Susan accessing it in the event of a divorce. This verbal argument escalated when Alex held Susan down on the ground and strangled her. This was witnessed by one of Susan’s older children. In the process of prying his fingers off her neck, Susan broke Alex’s finger. This stopped Alex from continuing to strangle her, but Susan was shocked, physically and psychologically hurt, and feared for her life. Susan grabbed a steak knife that was in close proximity on the kitchen bench and stabbed him in the shoulder. When the police arrived, Susan was as calm and collected as possible, and she had faith in the system protecting her, given she had been acting in self-defence. Susan was arrested, taken to the police station, where she was denied bail, and held overnight. The last thing her son said to her before she was taken away by police was that he didn’t want to be taken away from her.
Susan applied again for bail at court, explaining that she was defending herself, had no previous criminal history and would lose her house, job and kids into foster care if she didn’t get bail. The magistrate determined that these were not compelling reasons to grant bail, and Susan was held on remand. Susan, who had been working within the criminal justice system 24 hours earlier found herself on a prison transport vehicle headed for the state’s maximum security prison for women.
During her time in custody, Susan only saw her youngest child, who had spent his entire life in his mother’s care, twice and had no phone or visitation access with her other children. In consultation with her lawyer, Susan decided to plead guilty because she did not want experience or have her children experience having to testify in court. She received a custodial sentence and a community corrections order (CCO). She had already been in custody for the majority of the sentence she received so was released shortly after sentencing.
Upon her release, Susan was homeless. She received short-term crisis accommodation, and subsequently lived in shared accommodation, while rebuilding her life. She completed her CCO without issue. She lost her job due to the admission of guilt, and now has a criminal conviction and a job dismissal that will remain with her. Susan now has a new job, a house and one of her older children living with her. She knows she is lucky in this regard. She is entitled to visits with her other children on the weekends, however Alex continues to exert his control by withholding these visits frequently if Susan does not behave the way Alex wants or does not comply with his demands. This wasn’t just a case of misidentification, Susan was deemed the “only aggressor”, with Alex’s potentially lethal violence not acknowledged at all. This label and the ramifications of that night will stay with her for the rest of her life.
Source: Family Violence Reform Implementation Monitor, based on a direct victim survivor account
The Family Violence Report is designed to capture the predominant aggressor and person in need of protection (these terms are not used in the template but perhaps should be) and assess the risk of future harm to the victim(s). It also captures details of the family violence incident in the ‘narrative’ section and prompts referrals to Child Protection and specialist victim and perpetrator services. When officers complete the Family Violence Report on their device, they create linked ‘sub-incidents’ for any crimes that have occurred at the incident, which could include property damage or physical assault.34
There is no requirement for the respondent on the Family Violence Report to be the same as the offender in a crime. For example, a victim survivor who commits a crime in the process of exercising violent resistance can reasonably be labelled the offender in the sub-incident while still being acknowledged as the person in need of protection on the Family Violence Report. However, we understand that in practice police will usually match the respondent and offender, even if they know the offender is the victim survivor within the relationship. This is deeply problematic and is inconsistent with the intention of the Family Violence Report, MARAM Framework and police guidance about predominant aggressors. It sets both parties on the wrong justice and specialist service pathways.
For children using or allegedly using violence, many of whom have been living in family violence environments, The Orange Door practitioners, Centre for Excellence in Child and Family Welfare and Youthlaw agreed that they should not be labelled perpetrators or respondents. Instead, there should be a more trauma-informed approach, as has been implemented for child sexual abuse cases. This is consistent with our findings regarding adolescents using violence in the home in our 2020 report.35 We suggest that as whole-of-government efforts to respond to adolescent violence in the home progress, special consideration should be given to the risk of children under the age of 18 being misidentified as respondents [relates to action 16].
There is currently no explicit guidance for police in situations where a victim has technically committed a crime, and how to treat and record this. Victoria Police needs clear and specific guidance on how to approach these situations and to make a clear distinction between the criminal incident and the family violence risk assessment [relates to action 4]. Without practice improvement in this area, victim survivors are put in an impossible position in deciding whether to defend themselves. The victim survivor whose experience is shared in Susan’s story told us that she wonders if she should have let her ex-husband strangle her rather than fight back and suffer the consequences. Similarly, a practitioner from Elizabeth Morgan House, a specialist Aboriginal family violence service, said she wasn’t sure whether to advise women to defend themselves or not, knowing how easy it is for misidentification to occur when they do.
Internal review processes
Following initial police attendance at a family violence incident, review processes by supervisors should provide a crucial safeguard against misidentification being cemented in police records.
After a supervising sergeant approves the Family Violence Report, it is committed to LEAP, the police database. Victoria Police advised us that supervisors frequently return the brief (which includes the Family Violence Report, application for an FVIO and paperwork for any criminal charges) for review or changes to be made. It is not clear how often supervisors pick up misidentification or seek further information before approving the brief [relates to action 14]. However, many stakeholders noted that there is considerable room for improvement at this point in the process. Supervisors represent a smaller, more experienced part of the police workforce who can support the on-the-job learning of the younger general duties officers who are under considerable work pressure. Supervisor training in being able to accurately identify the person most in need of protection and the predominant aggressor, and to understand how to capture this in police records is, therefore, essential.
Internal family violence expertise can be sought from family violence liaison officers, who are sergeants with family violence as one of their portfolios, although these roles only exist at 24-hour stations. Family Violence Investigation Units are also a source of internal Victoria Police family violence expertise, but it is not clear how often advice is sought or how frequently this leads to misidentification being detected.
One recurring suggestion during our consultations was for a stronger partnership between Victoria Police and the family violence sector to review cases where there is a high risk of misidentification or other uncertainty about the dynamics in the relationship. For example, it may be worth considering a clear review process for cases where a woman is identified as a respondent, where the narrative talks about mutual violence, or for LGBTIQ+ relationships. This process could include using information sharing mechanisms to check past engagement with family violence and other relevant services, which would be particularly useful for cases where there are no past police incidents [relates to action 5]. This should ideally occur before the Family Violence Report is committed to LEAP and be seen as an opportunity to build capability and support improved practice. Such a model would be worth trialling to determine the level of demand pressure it would place on the sector, whether the review can occur within timeframes required by police, and whether it reduces the rate of misidentification. It will also be important to assess whether such a review process is equally benefiting marginalised groups who are overrepresented, particularly Aboriginal women [relates to action 14].
When the perpetrator is a police officer
The issue of perpetrators within Victoria Police has direct relevance to the issue of misidentification. Police officers who are perpetrators are likely to be less responsive to family violence training, less likely to properly investigate family violence, and may deliberately collude with other perpetrators. In other cases, a police perpetrator may deliberately cause misidentification. We spoke with one victim survivor who experienced family violence at the hands of a police officer who abused his power to have her labelled a perpetrator; her experience is captured in Sally’s story.
Among the Victorian population, there were 123 perpetrators involved in family violence incidents for every 10,000 people in 2019.36 The Age has reported that only 23 Victorian police officers (of the 16,295-person workforce, or 0.14 per cent) were charged with family violence offences in that same year.37 Evidence suggests this is probably an underestimate and that police are ‘as likely as the general public to be perpetrators of intimate partner violence’.38
Victoria’s Independent Broad-based Anti-corruption Commission is quoted as having concerns about ‘systemic problems in how Victoria Police handles matters where there are family violence allegations against a serving member’.39
Sally was in a relationship with a long-serving police officer, Tom. The relationship included physical abuse (but he was careful not to leave visible bruises), property damage, emotional abuse and threats against their pets. Due to his position, Sally was understandably apprehensive about going to police, so she had never approached police or family violence services for help. Tom had made it clear through his emotional abuse that if she did, she would not be believed anyway. He even told her that if she went to police, he would say she was the primary aggressor. She didn’t even know what that term meant.
Sally left Tom when she was pregnant with her second child. After leaving the relationship, the abuse morphed into coercive control from afar. Tom has used the children as leverage against her by way of family consent orders. In the Federal Circuit and Family Court of Australia, Tom has attempted to manipulate systems against Sally by claiming she has mental health issues, alcohol issues and has abused him, in attempt to discredit her character and make himself look favourable to the courts. When she doesn’t agree to his demands, he continues to use threats, punishment and coercion to get his way. Tom has exercised misidentification as his tool of trade, using victims’ words he has heard during the course of his work as a police officer to portray himself as the victim. In Sally’s words, Tom is seen as charming and unassuming – people like him and want to believe him. Sally eventually agreed to equal shared custody of the children to end the traumatic court process.
Last year, Sally was served with an application for an FVIO, on which she was listed as the respondent. She knew nothing of the order until police arrived at the door to serve her. The police officers were Tom’s colleagues. This led to a lengthy, distressing and retraumatising court process. Tom’s role, knowledge of the system and perceived credibility has meant that Sally has constantly felt as though it was ‘her against the police’. Sally sought legal support and the order was later revoked when the police agreed to voluntarily withdraw. But this was not before Sally had to tell her story again and describe him as the perpetrator.
Due to their shared custody arrangements, Sally constantly feels unsafe, waiting for Tom’s next attack and terrified of his threats to exert further control over her via the children. She feels she is continually defending herself and fighting to be believed, which deliberately deflects away from his wrongdoing. She believes that her children are also feeling the lifelong impact and she despairs at the knowledge that her children must endure his abuse when they are in his household every fortnight.
Source: Family Violence Reform Implementation Monitor, based on a direct victim survivor account
Having perpetrators within Victoria Police weakens the implementation of its efforts to effectively respond to family violence. It is encouraging that Victoria Police is actively working to address this, with the Victoria Police Manual Family Violence section recently released on ‘Family violence involving Victoria Police employees’ and the first of two employee family violence teams of specialist detectives commencing in November 2021.
27 This report focussed on the third stage of the Multi-Agency Triage (MAT) project. Between July and December 2016 and July and
November 2017, a total of 7462 police Family Violence Reports were received by Berry Street Specialist Family Violence Service in the
north-eastern region of Melbourne. These were triaged daily by a multi-agency team and appropriate risk assessment and service
response was determined.
28 Humphreys, C. & Nicholson, D (2017): Multi-Agency Triage Project: Final Report. Melbourne, University of Melbourne, p.13. Available at: safeandtogetherinstitute.com/wp-content/uploads/2018/06/MAT-Final-Report_6Feb_2018-2.pdf.
29 For example: The Age, 18 May 2021: ‘Rosie told police she was a victim of domestic violence. She was the one arrested’. Available at: abc.net.au/news/2021-05-19/rosie-told-police-victim-of-domestic-violence-she-was-arrested/100112832 (accessed 7 July 2021); Guardian Australia, 20 September 2021: ‘‘Denied a voice’: how Australia fails migrant victims of domestic violence’. Available at: theguardian.com/australia-news/2021/sep/20/denied-a-voice-how-australia-fails-migrant-victims-of-domestic-violence (accessed 20 September 2021).
30 Jamieson LM, Harrison JE, Berry JG (2008): Hospitalisation for head injury due to assault among indigenous and non-indigenous Australians, July 1999–June 2005, The Medical Journal of Australia, 188(10): pp. 576–579.
31 Victorian Equal Opportunity and Human Rights Commission (2019): Independent review into sex discrimination and sexual harassment, including predatory behaviour, in Victoria Police: Phase 3 audit and review, p. iv. Available at: humanrights.vic.gov.au/static/9a9be4301421aec42a6cb3620eca1f01/Resource-Independent_Review_Victoria_Police-Phase_3-2019.pdf (accessed 21 September 2021).
32 Crime Statistics Agency (2021): Criminal incidents and rate per 100,000 population by principal offence and family incident flag. Available at: crimestatistics.vic.gov.au/crime-statistics/latest-victorian-crime-data/recorded-criminal-incidents-2 (accessed 22 September 2021).
33 For further information on ‘collusion’, see Family Safety Victoria (2021): MARAM practice guides: Foundation Knowledge Guide, p. 47. Available at: (accessed 21 October 2021).
34 If sexual assault is disclosed, officers will contact their Sexual Offences and Child abuse Investigation Team.
35 Family Violence Reform Implementation Monitor (2021): Report of the Family Violence Reform Implementation Monitor as at 1 November 2020. Available at: .
36 Family Violence Reform Implementation Monitor calculations based on data from the Crime Statistics Agency, Latest Victorian crime data – Family incidents – rate per 100,000 population. Available at: crimestatistics.vic.gov.au/crime-statistics/latest-victorian-crime-data/family-incidents-2 (accessed 20 September 2021).
37 ABC News (2020): 'Abusers in the ranks', 19 October 2020. Available at: (accessed 21 October 2021).
38 Larsen AC, Guggisberg M (2009): Police Officers, Women and Intimate Partner Violence: giving primacy to social context, p.12. Available at: (accessed 21 October 2021).
39 IBAC, quoted in The Age (2020): ‘‘Systemic problems’: IBAC uncovers police failings on domestic abuse by officers’, 6 December 2020. Available at: (accessed 24 September 2021).
Reviewed 16 December 2021