Family violence perpetrators: Existing evidence and new directions

Dr Bronwyn Morrison
Principal Research Adviser, Department of Corrections
John Davenne
Senior Adviser, Service Development, Department of Corrections

Author biographies:
Bronwyn Morrison has a PhD in Criminology from Keele University, UK. During the past 10 years she has worked in research roles across government, including NZ Police, Ministry of Justice, and Department of Conservation. She joined Corrections in March 2015.

John Davenne has a BA Hons in Probation & Community Justice from Birmingham University, UK. During the past 15 years he has worked in a variety of probation, programmes and criminal justice system roles, both in the UK and NZ. He joined Corrections in March 2011.


Responding to family violence consumes a significant amount of government resource, particularly in relation to the criminal justice system. In 2014, NZ Police commenced more than 100,000 family violence investigations. Over 9,200 family violence-related prosecutions were processed through New Zealand courts in 2014, and over 5,100 applications for protection orders were lodged, with over 3,100 final orders granted. Just under one fifth of all sentences managed by the Department of Corrections in 2014 included at least one family violence-related offence, and 6,212 individuals started sentences where the most serious offence was family violence. Despite the significant resource implications associated with these volumes, research evidence on the nature and extent of family violence in New Zealand remains limited. Crucially there has been little research conducted on those who perpetrate family violence, including the specific contexts, situational dynamics, and desistance processes associated with their violence. There has also been little evaluative work within New Zealand to identify the most effective ways of responding to family violence perpetrators, either within the context of the criminal justice system or outside of it.

While more recent years have seen a greater focus on family violence victims, both within criminal justice policy and research, perpetrators have not attracted much attention. While it is widely agreed that family violence perpetrators should be ‘held to account’, there is little consensus about what this actually means in practice, nor whether ‘being held to account’ is sufficient to either stop (or at least reduce) the perpetration of family violence. Internationally there has been a small, but growing, concession that in order for responses to ‘work’ they need to do so for both victims and perpetrators (Centre for Innovative Justice, Australia, 2015). Following these international developments, the recent Ministerial Review of Family Violence and Sexual Violence aimed to shed further light on family violence perpetrators in New Zealand. As part of the review, the Department of Corrections led an interagency work stream on follow-up and long term responses to perpetrators. This included a review of extant evidence on family violence perpetrators in New Zealand and internationally, alongside a critical assessment of best or, more often, emergent practice in perpetrator responses. A review of current service provision for family violence perpetrators in New Zealand was simultaneously completed.

This paper presents a brief summary of the main findings from the review, highlighting evidence gaps and directions for future research and service innovation. While the focus of the Ministerial Review was very broad, the focus of this paper is predominantly on Intimate Partner Violence (IPV).

What is known about perpetrators of intimate partner violence?

To date, studies on family violence perpetrators in New Zealand have often been small scale and/or based on non-representative samples (see, Roguski & Gregory, 2014; King, 2011, Department of Corrections, 2015). Consequently, existing knowledge about those who commit family violence offences in New Zealand is limited, especially in relation to those who do not enter the criminal justice system. This is an important caveat to bear in mind when interpreting the findings presented below. As international studies have aptly demonstrated, research based on different samples of offenders can give rise to varying findings about who perpetrates family violence and the nature of that violence (Johnson, 2008; Kelly & Johnson, 2008; Dutton, 2006). For example, studies based on surveys of the general population offer different results to those based on samples identified through women’s shelters or correctional settings, with the former reporting much greater levels of gender symmetry ,as well as lower frequency and intensity of violence (Johnson, 2008; Babcock, Robie & Green, 2004).

Despite the frequent assertion that IPV is not class specific, research has consistently shown that the majority of family violence perpetrators dealt with within the criminal justice system come from backgrounds characterised by multiple, sometimes intergenerational, disadvantage. IPV perpetrators in this context tend to present with problems linked to unemployment, poverty, substance misuse/addiction, and mental health conditions, and often have extensive criminal offending histories (Slabber, 2012; Gondolf, 2004, 2012; Gray et al., 2014; Boxhall, Payne & Rosevear, 2015; Gadd, 2004). However, these histories do not always include prior convictions for family violence offences (Boxall et al., 2015). Of those who perpetrated IPV homicides in New Zealand in 2013, for example, only 11% were previously known to the Police as family violence offenders, while about one fifth of those starting sentences for family violence in 2013 appeared to have no history of family violence offending (Department of Corrections, 2015).

The links between family violence, sexual violence and child abuse have often been noted in political and public discourse on family violence. However, although many of those entering the criminal justice system for family violence have a history of committing other offences, evidence suggests family violence perpetration does not always coincide with sexual violence and/or child abuse. Only a small proportion (6%) of those sentenced for family violence offences in 2013 were charged with sexual offences; just under one third of those starting sentences for sexual offences had committed the offence against a family member. In other words, the majority of those sentenced for sexual offences in 2013 were convicted for offences that did not coincide with family violence (Department of Corrections, 2015). While there is some overlap, and undoubtedly a high level of under-reporting, the relationship between sexual and family violence should not be automatically assumed.

The overlap between IPV and child abuse appear to be stronger than that found for sexual abuse. Although assessments vary due to different definitions, it has been estimated that 30% to 60% of child abuse occurs in households where IPV is co-occurring (Barnish, 2004). Australian research has found that child abuse is 15 times more likely in households where IPV is occurring, while UK research indicates that one in three child protection cases and 40% of child sexual abuse cases involved co-occurring IPV (Barnish, 2004). It has also been estimated that around 50% to 60% of prisoners attending family violence programmes either witnessed or directly experienced domestic violence as children, and that up to 40% of those abused as children go on to perpetrate IPV (Stewart, Gabora, Kropp & Lee, 2014). Like sexual violence, the relationship between child abuse and IPV is an imperfect one and it is important to note that the majority of those abused as children do not go on to perpetrate IPV.

It is widely agreed that men are more likely to be the primary aggressor in IPV cases that result in prosecution, and are far more likely than women to engage in chronic, repeat and severe IPV, which is coercively controlling and fear-inducing (Swan et al., 2008; Johnson, 2008). Looking at family violence perpetrators managed by Corrections, it is evident that women make up a small proportion of IPV perpetrators in New Zealand (10% in 2013); however, the number of women serving sentences for IPV-related offences and/or subject to protection orders is increasing. In terms of ethnicity, international evidence conclusively shows that ethnic minority groups are over-represented both as victims and offenders of IPV. This holds true in New Zealand. For example, analysis based on a family violence Corrections cohort from 2009/10 reveals that over half (51%) were Mäori, and 12% identified as Pacifica.

Family violence perpetrators also exhibit considerable heterogeneity. In addition to the factors noted above, a number of international studies have highlighted different patterns of violence committed by family violence perpetrators (Johnson, 2008; Kelly & Johnson, 2008; Dutton, 2006). Four key patterns of family violence, in particular, may be useful for thinking about the ways in which violence functions in intimate relationships, namely: situational couple violence (that which does not exhibit power and control dynamics); violent resistance (violence committed by, typically female, victims against a primary (often male) aggressor); separation violence (violence commencing at the end of a relationship in which violence was not previously present); and, finally, coercive control (fear-inducing psychological and emotional intimidation and/or physical violence). To date there has been no research conducted on the incidence or prevalence of different types of IPV in New Zealand, although policy and research is often produced as if coercive control is either the prevalent or only form of violence. Further research undertaken to obtain a better understanding of the nature and stability of different types of IPV would be invaluable. Such knowledge has the potential to better direct IPV treatment and desistance.

While there has been considerable research focused on measuring the effectiveness of group programmes, there has been comparatively little exploring family violence desistance. This is true in New Zealand, where there is an absence of research on family violence desistance and its relationship with general criminal desistance. The small amount of international research which has explored this issue indicates that desistance from family violence reflects the crime-age curve found for general offending, with most perpetrators ‘aging out’ of family violence (Capaldi & Kim, 2007; Walker, Bowen, Brown & Sleath, 2014).

Evidence indicates that the type of violence and perpetrator matters to desistance, with those predominantly involved in situational couple violence showing a greater proclivity to desist compared to coercive controlling perpetrators and/or those with significant personality disorders, mental health problems, substance abuse/dependence issues, and extensive criminal histories (Walker et al, 2015; Walker, Bowen & Brown, 2013; Johnson, 2008; Kelly and Johnson, 2008). Other factors associated with desistance include: being ‘held to account’ and recognising the impact of negative abusive behaviours; the development of a non-violent identity; coming to terms with experiences of childhood trauma, economic marginalisation and a sense of powerlessness; strong external support networks; and acquiring self-regulation skills, such as techniques for managing emotional triggers (Walker et al 2014; Morran, 2010 2013; Giordano et al., 2015; Roguski & Gregory, 2014).

Western treatment models for family violence perpetrators are predominantly typified by one of two approaches: the Duluth Model or Cognitive Behavioural Therapy (CBT); although in practice many programmes blend the two. The Duluth model, developed in Minnesota, USA in the 1980s, represents a feminist psycho-educational approach predicated on a belief that men’s violence against women is the product of power-control dynamics in society more broadly. This model has been traditionally endorsed as best practice for family violence perpetrator programmes. However, following three decades of implementation and extensive research investment, there is little conclusive evidence to demonstrate the model is beneficial in practice (Day et al., 2010; Babcock, Green and Robie, 2004). Questions have arisen about the ability of the Duluth model to address different varieties of family violence, including same sex violence and women’s use of intimate partner violence, both as sole perpetrators and mutual combatants (Dutton & Corvo, 2007). Recognition of this limitation, coupled with growing international accord on ‘what works’ with offender treatment more broadly, has seen a gradual shift toward more CBT-based modalities in many jurisdictions.

CBT-based programmes are underpinned by the notion that an individual’s disposition and actions can be improved by identifying and replacing dysfunctional thinking (Slabber, 2012; McGuire, 1996) and through acquiring adaptive cognitive and interpersonal skills. CBT presumes that violence stems from learnt behaviours, which can be modified or replaced with new behaviours. For example, the reduction of anger can occur through learning emotional concepts, taking time out/in, utilising relaxation techniques, learning and practicing negotiation skills, and developing the ability to tolerate unpleasant and/or strong emotions.

Alongside the greater utilisation of CBT, more recent research has made a compelling case for integrating the principles of Risk-Needs-Responsivity ( ) into family violence programmes (Radatz & Wright, 2015; Stewart et al. 2014; Grealy et al., 2012). These principles, internationally proven to enhance the effectiveness of interventions for general offending, have been increasingly found to work for family violence offenders also (Andrews & Bonta, 2010; Radatz & Wright, 2015).

Briefly put:

  • The Risk principle holds that interventions should be commensurate with an individual’s assessed risk; the higher the risk, the more intensive the treatment.
  • Need puts forward the notion that ‘criminogenic’ needs, those dynamic factors associated with recidivism, should be targeted.
  • Responsivity involves two parts which are both specific and general: the specific entails taking into account the individual’s personal aspects and tailoring treatment to more effectively facilitate learning. For example, paying attention to internal and external barriers, learning and thinking styles, mental disorders, and personality types.

In line with growing international evidence, the Department of Corrections has recently re-designed treatment pathways for family violence perpetrators. As a consequence, low to low-medium risk male offenders are directed to the Family Violence Programme (run internally by Corrections staff as well as by contracted non-government organisations). Medium risk male offenders are potentially eligible for either the Medium Intensity Rehabilitation Programme (MIRP) or the Short Rehabilitation Programme (SRP-M). High risk perpetrators are directed to Psychological Services for individual treatment. Female offenders are eligible for referral to Kowhiritanga (equivalent to the MIRP) and the Short Rehabilitation Programme for Women (SRP-W).

In late 2014 the Department of Corrections piloted the Family Violence Programme, which incorporates both CBT and the RNR principles. Similarly, the Ministry of Justice introduced a new Code of Practice for their family violence programme providers in 2014, which broadly endorses these approaches, albeit less prescriptively than Corrections. While insufficient time has passed to measure the implications of these changes, this will likely be a future research focus.

In addition to programme content, evidence has suggested that other factors may be as, if not more, important. For example, Miller, Duncan & Hubble (2004) found that 40% of treatment outcomes were related to extra-therapeutic factors (such as social supports, client skills and individual motivation), 30% pertained to the client-therapist relationship, 15% to the facilitators’ ability to impart a sense of hope to clients, and 15% to the specific programme content (cited in Cagney & McMaster, 2013). Whether programme content is delivered as intended (programme integrity) and how programmes are facilitated on the ground will therefore remain key considerations for future service development, monitoring, and evaluation activity. Finally, as widely attested internationally, perpetrator programmes have tended to function most successfully when they are either integrated, or at the very least aligned, with services for victims and, increasingly, any children from the relationship. Internationally, integration has also increasingly included combining programme provision with “wrap-around” support services for perpetrators. Defining precisely what form such integration could or should take in New Zealand is still under development.

In summary, notwithstanding decades of government investment and public interest, knowledge about family violence perpetrators in New Zealand and how best to respond to this group merits further development. That said, there has been a welcome movement away from viewing responses dichotomously as either for victims or for offenders. It is now widely agreed that understanding and effectively responding to perpetrators is crucial for improving outcomes for victims. There has been an increased willingness to step outside of the ‘nothing works’ doctrine historically attached to perpetrator programmes to more fully consider what it means for responses to ‘work’, for whom they work best, when, and why, and how effectiveness can be improved and reliably measured. There is still much work to be done; however, as New Zealand embarks on a range of family violence innovations, including the national availability of the Department of Correction’s Family Violence Programme, there is cause for (cautious) optimism. While family violence is an ingrained, “wicked” problem requiring ongoing investment and sustainable funding, family violence perpetrators can and do stop offending, and there is opportunity for the Department of Corrections to make a meaningful contribution to the desistance process.

References
Andrews, D., and Bonta, J. (Eds.). (2010). The Psychology of Criminal Conduct (5th Edition ed.). New Providence, NJ: Matthew Bender & Company, Inc., LexisNexis Group.

Babcock, J.C., Green, C.E. and Robie, C. (2004) ‘Does batterers’ treatment work? A meta-analytical review of domestic violence treatment’, Clinical Psychology Review, 23: 1023-1053.

Barnish, M. (2004) Domestic Violence: A Literature Review. London: HM Inspectorate of Probation.

Boxall, H., Payne, J., and Rosevear, L. (2015) ‘Prior offending among family violence perpetrators: a Tasmanian sample’, Trends and Issues in Crime and Criminal Justice, no.493. Canberra: Australian Institute of Criminology.

Cagney, M. and McMaster, K. (2013) ‘Men’s Intervention Programs’, DVRCV: International Interventions, 1. Autumn/Winter, 13-17.

Capaldi, D.M. and Kim, H. K. (2007) ‘Typical approaches to violence in couples: a critique and alternative conceptual approach’, Clinical Psychology Review, 27, 253-265.

Centre for Innovative Justice (2015) Opportunities for early intervention: Bringing perpetrators of family violence into view. Melbourne: RMIT University.

Day, A., Chung, D., O’Leary, P., Justo, D., Moore, S., Carson, E. Gerace, A. (2010) ‘Integrated responses to domestic violence: mandated intervention programmes for male perpetrators’, Trends and Issues in Crime and Criminal Justice, 404. Canberra: Australia Institute of Criminology.

Department of Corrections (2015) Family violence offenders, Topic Series. Wellington: Department of Corrections.

Dutton, D. (2006) Rethinking Domestic Violence. Vancouver: UBC Press.

Dutton, D. and Corvo, K. (2007) ‘The Duluth Model: A data impervious paradigm and failed strategy’, Aggression and Violent Behaviour. 12, 658-667.

Gadd, D. (2004) ‘Evidence-led policy or policy-led evidence? Cognitive behavioural programmes for women who are violent towards women’, Criminal Justice, 4 (2): 173-197.

Giordano, P., Johnson, W.L., Manning, W,D., Longmore, M.A., and Minter, M.D. (2015) ‘Intimate partner violence in young adulthood: Narratives of persistence and desistance’, Criminology: 1-36.

Gondolf, E. W. (2012) The Future of Batterer Programmes: Reassessing Evidence-Based Practice. New England: Northeastern University Press.

Gondolf, E. W. (2004) ‘Evaluating batterer counselling programs: A difficult task showing some effects and implications’, Aggression and Violent Behaviour, 9: 605-631.

Gray, R., Broady, T., Gaffney, I., Lewis, P., Mokany, T., and O’Neill, B. (2014) “I’m working towards getting back together’: Client accounts of motivation related to relationship status in men’s behavioural change programmes in New South Wales, Australia’, Child Abuse Review, DOI 10.1002/car.2318.

Grealy, C., Wallace, A., Wilczynski, A., Lai, S., Bodiam, T., Dowler, B. and Jones, L. (2012) Literature Review on Domestic Violence Perpetrators. Melbourne: Urbis.

Johnson, M. P (2008) A Typology of Domestic Violence: Intimate terrorism, violent resistance, and situational couple violence. Boston: Northeastern University Press.

Kelly, J. B. and Johnson, M.P. (2008) ‘Differentiation among different types of intimate partner violence: research update and implications for interventions’, Family Court Review, 46(3): 476-499.

King, L. (2011) Community-Based Domestic Violence Programmes: A Review of the Adequacy of Programmes Available. Wellington: Department of Corrections.

McGuire, J. (Ed) (1996) What Works: Reducing Reoffending. Guidelines from Research and Practice. John Wiley & Sons, Chichester.

Morran, D. (2013) ‘Desistance from domestic abuse: Influences, patterns and processes in the lives of formerly abusive men’, The Howard Journal of Criminal Justice, 52 (3): 306-320.

Morran, D. (2010) ‘Re-education or recovery? Re-thinking some aspects of domestic violence perpetrator programmes’, The Journal of Community and Criminal Justice, 58 (1), 23-36.

Radatz, C.L. and Wright, E. M (2015) ‘Integrating the principles of effective intervention into batterer intervention programming: the case for moving towards more evidence-based programming’, Trauma, Violence and Abuse, July, 1-16.

Roguski, M. and Gregory, N. (2014) Former family violence perpetrators narratives of change. Auckland: The Glenn Inquiry.

Slabber, M. (2012) Community-Based Domestic Violence Interventions: A literature review. Wellington: Department of Corrections.

Stewart, L.A., Gabora, N., Kropp, P.R. and Lee, Z. (2014) ‘Effectiveness of Risk-Needs-Responsivity-based family violence programmes with male offenders’, Journal of Family Violence, 19: 151-164.

Swan, S.C. and Gamborne, L.J. Caldqwell, J. E. Sullivan, T. P., and Snow, D. L. (2008) ‘A review of research on women’s use of violence with male intimate partners’, Violence Vict. Melbourne: National Institute of Health.

Walker, K., Bowen, E., Brown, S. and Sleath, E. (2015) ‘An examination of psychopathy among men who have suspended the use of family violence in their intimate relationships’, Journal of Family Violence, 30, 539-554.

Walker, K., Bowen, E., Brown, S., and Sleath, E. (2014) ‘Desistance from intimate partner violence: a conceptual model and framework for practitioners for managing the process of change’, Journal of Interpersonal Violence, 1-25.

Walker, K., Bowen, E., and Brown, S. (2013) ‘Desistance from intimate partner violence: a critical review’, Aggression and Violent Behaviour, 18, 271-298.