Australian Institute of Criminology

Skip to content

Partner violence among a sample of police detainees

Trends & issues in crime and criminal justice no. 337

Jenny Mouzos and Lance Smith
ISBN 978 1 921185 41 0 ; ISSN 0817-8542
Canberra: Australian Institute of Criminology, May 2007

Foreword | Using data from the Drug Use Monitoring in Australia (DUMA) program, this paper explores involvement in intimate partner violence, and provides first-time results from face-to-face interviews with a group of 1,597 police detainees. The study found that the levels of intimate partner violence are much higher among this group (49%) than is found from general population surveys. More than two-thirds of the detainees who were involved in partner violence reported being both a victim and a perpetrator in the past 12 months. This is consistent with other criminological research that finds that a large proportion of offenders often report having been victims of crime. Factors found to be significant risk markers for detainees' involvement in partner violence included prior arrest, drug and alcohol dependency, having dependent children, and experiencing physical abuse as a child. Although a greater percentage of female than male detainees reported being involved in partner violence, once these other risk factors were controlled for, gender was not found to be a significant risk marker among this sample. The paper discusses the policy implications of the findings, particularly the need for early intervention with at-risk families and where drug and alcohol dependency issues are emerging.

Toni Makkai
Director

As violence in the home has moved from being considered largely a private issue into the public sphere, it has attracted considerable attention. While much of this has focused on violence directed at women, a number of recent research initiatives examined violence against both men and women (ABS 2006; Johnson 2004; Mouzos & Houliaras 2006). Most of this research sought to determine the prevalence and/or incidence of violence experienced by (predominantly) women and men among the general population. In Australia, during the 12 months prior to the survey, 16 percent of women and six percent of men experienced physical assault from a current partner (ABS 2006). This increased to 22 percent for women and 20 percent for men for experiences of violence from a former partner during the 12 months prior to the survey (ABS 2006). If incidents of extreme violence, such as homicide are considered, then men predominate as perpetrators (83%). When women commit homicide, they are most likely to kill an intimate partner or a family member (66%; Mouzos & Houliaras 2006).

International research finds levels of violence directed against women and men to be similar. For example, results from the 1999 General Social Survey on Victimisation in Canada reported that over the previous five years, eight percent of women and seven percent of men reported experiencing at least one incident of relationship violence (Statistics Canada 2003). Other research found that the prevalence of partner violence was higher for distinctive groups, such as women in prison (e.g. Johnson 2004). A study of incarcerated female offenders in six jurisdictions in Australia during 2003 found that 78 percent of the women interviewed reported experiencing at least one type of abuse (emotional, physical or sexual) in adulthood. Just over half of the women in this study experienced physical abuse by a spouse/partner over their lifetime (Johnson 2004).

While violence directed against women by intimate partners has long been recognised, violence against men is still considered to be a hidden social problem (Cook 1997). Some suggest there is a need to move away from viewing domestic violence as a gender issue with males as perpetrators and females as victims, and consider it as involving couples who are violent towards one another (Fergusson, Horwood & Ridder 2005). Others argue that only some types of intimate partner violence are gender symmetric (Johnson 2005).

The gender paradox seems to arise when prevalence rates in differing populations are examined. As noted earlier in this paper, estimates from general population surveys found similar rates of intimate partner violence over the lifetime for men and women (see ABS 2006), whereas prevalence rates derived from police data (particularly homicide offence records; Mouzos & Houliaras 2006), refuges or emergency department data indicate that levels of intimate partner violence victimisation are higher for women than men (Fergusson, Horwood & Ridder 2005).

The contentious gender issue in intimate partner research suggests a need for further research. There are also relatively few studies on intimate partner violence drawn from an at-risk arrestee population.

Purpose of the current study

The purpose of this paper is to examine intimate partner violence among a sample of police detainees. Associations with their drug use and other offending behaviour are also examined. Most importantly, this paper undertakes a comparative analysis based on the gender of the police detainees, providing information on both male and female perpetration and victimisation of partner violence. The risk markers for involvement in partner violence are also examined.

Methodology

The paper examines the issue of intimate partner violence among a sample of adult men and women who are detained by police in a police station or watchhouse involved in the Drug Use Monitoring in Australia (DUMA) program.

In brief, the DUMA program now operates across nine sites in Australia on a quarterly basis, and involves both voluntary interviews with police detainees, and urinalysis (see Makkai 1999 for additional information on the methodology). In addition to the questionnaire which incorporates standard questions on drug use and offending behaviour, a different addendum questionnaire is administered to the detainees each quarter asking about topics of specific concern (e.g. drug driving, amphetamines, weapons). In the first quarter of 2003 and the third quarter of 2005, an addendum on violence in the home was administered to detainees in seven of the sites.

The addendum sought to determine the prevalence and incidence of partner violence experienced and perpetrated by police detainees over their lifetime and during the past 12 months, as well as incidents witnessed or experienced as a child in the home. In total, 1,597 police detainees (1,355 males; 242 females) provided details of their experiences of intimate partner violence. A comparative analysis of the two quarters in which the addendum was run showed few differences between the quarters so the data from both quarters were pooled.

For the purposes of the addendum, a partner was defined as a person with whom the individual has or had an intimate, romantic, or sexual relationship, whether they lived with the person or not. This includes: a girlfriend/ex-girlfriend, boyfriend/ex-boyfriend, spouse/ex-spouse.

Prior to discussing the results of the addendum, a number of issues need to be highlighted. First, self-report surveys rely on the recall of events by the participants. The accuracy of this information is thus dependent on the respondents' ability to recall events correctly, as well as their willingness to provide candid and honest answers (see Skogan 1986 for an overview of factors that may influence the accuracy of responses provided in crime victim surveys).

These issues are further exacerbated in relation to self-reported drug use and offending behaviour. Research that examined the prevalence of under-reporting of drug use among police detainees found that, while the majority of detainees accurately reported their drug use (McGregor & Makkai 2003), there were some differences in the characteristics of those detainees who were more likely to under-report their recent drug use. For example, detainees who used illicit drugs and were in frequent contact with the criminal justice system or in treatment were more likely to accurately report, whereas detainees who led more socially acceptable lifestyles were more likely to under-report their drug use (McGregor & Makkai 2003).

Another point to note is that the study population is police detainees, a specific subgroup of offenders - those arrested by police. The results may not be applicable to the whole offender population, nor to the general population as a whole.

Experiences of partner violence

A higher percentage of female than male detainees reported experiencing at least one incident of a physical confrontation with an intimate partner over their lifetime (81% versus 56%; Table1).

Table 1 : Lifetime experiences of partner violence, by gender (percent)
Male Female
* 2 test significant at p
Source: AIC, DUMA collection 2003-2005 [computer file]
Ever had a physical fight with an intimate partner* 38 75
Victim of a physical confrontation (hit/pushed/shoved/slapped)* 56 81
Perpetrator of a physical confrontation* 37 51
Spouse or partner has been arrested for an incident related to domestic violence* 7 39
Ever been arrested for an incident related to domestic violence* 22 13
Ever been involved in partner violence (as victim or perpetrator)* 62 85
Total (n) (1,355) (242)

Of note, however, is that the level of reported violence is markedly higher among both male and female detainees than among the general population.

Female detainees were more than five times more likely than male detainees to report that their intimate partner had been arrested for an incident related to domestic violence (Table1). Similarly, male detainees were more likely than female detainees to report that they themselves had been arrested for an incident related to domestic violence (22 percent versus 13 percent).

The pattern is similar for intimate partner violence experienced during the previous 12 months, with female detainees reporting higher levels than male detainees (Table2). Overall, half of the female detainees and over four in ten of the male detainees reported experiencing at least one incident of partner violence victimisation during the past 12 months (Table2).

Table 2 : Experiences of victimisation by an intimate partner during the past 12 months, by gender (percent)
Type of violenceMale Female
* 2 test significant at p
Source: AIC, DUMA collection 2003-2005 [computer file]
Threatened to hurt you physically in a way that frightened you* 13 41
Thrown something or hit you with something that could hurt* 28 36
Pushed or grabbed you or twisted your arm or hair pulled in a way that hurt or frightened you* 20 45
Slapped, kicked, bit, or hit you with a fist 30 37
Used a knife or gun used on you* 8 15
Experienced any partner violence (victim only)* 41 53
Total (n) (1,011) (199)

For some types of violence such as 'threatened to hurt you physically in a way that frightened you', there was considerable disparity, with only 13 percent of males reporting being threatened, compared with 41 percent of females. Some other types of violence did not involve such large differences. For the category of 'slapped, kicked, bit or hit you with a fist' males were only slightly less likely than females to have experienced this type of violence in the past 12 months (30% and 37% respectively). This was the only type of violence where the difference between the genders was not statistically significant.

While a significantly higher percentage of female detainees reported being victimised by their intimate partners, a statistically significant higher percentage of females also reported perpetrating intimate partner violence during the past 12 months than male detainees (47% compared to 38%; see Table3).

Table 3 : Perpetration of partner violence during the past 12 months, by gender (percent)
Type of violenceMale Female
* 2 test significant at p
Source: AIC, DUMA collection 2003-2005 [computer file]
Thrown something at a partner that could hurt* 11 26
Twisted your partner's arm or pulled their hair 13 11
Pushed or shoved your partner* 28 37
Grabbed your partner 26 27
Slapped your partner* 12 27
Used a knife or a gun on your partner* 1 8
Punched or hit your partner with something that could hurt* 7 22
Beat up your partner* 3 10
Kicked your partner* 3 18
Perpetrated any of the above acts of partner violence* 38 47
Total (n) (1,008) (201)

In terms of particular types of violence, women were significantly more likely to admit throwing something at, pushing or shoving, slapping, punching or kicking their partner. However, there were some types of behaviours for which there were no gender differences. For example, about a quarter of the male and female detainees reported having grabbed their partner, while about 10 percent reported twisting their partner's arm or pulling their hair (Table3). Few detainees reported having used a knife or gun on their partner.

Overall, just over half (51%) of the police detainees who participated in the addendum survey reported that they had neither experienced or perpetrated any partner violence in the past 12 months.

However, of the detainees who were involved in partner violence (n=589), 68 percent reported being both a victim and a perpetrator in the past 12 months. Gender differences were also found in this group. Two-thirds of the male detainees (66%) and around three-quarters of the female detainees (74%) reported being both victims and perpetrators of partner violence. The percentage of female detainees who reported being victims only was similar to that of male detainees (20% versus 19%). In contrast, males were more likely than female detainees to report being perpetrators only (14% versus 7%). However, it should be noted that the number of detainees in each of these groups was quite small (victim only n=117; perpetrator only n=73).

Given the large overlap between victimisation and perpetration of partner violence in the sample of police detainees who participated in the addendum, as well as the small sample sizes when examining victimisation and perpetration separately, it was decided to combine the three groups of detainees (both victim and perpetrator, victim only, and perpetrator only) into one group (victim and/or offender of partner violence) for all subsequent analyses. The aim was to focus on those detainees who had been involved in partner violence compared with those detainees who had not been involved.

Risk markers for partner violence: the role of alcohol, drugs and offending

This section examines the sociodemographic characteristics of detainees who were involved in partner violence. It also explores the role of alcohol, drugs and crime as risk markers for partner violence. Do detainees who were involved in partner violence as a victim and/or perpetrator also engage in higher levels of substance use and other crime than detainees who have not been exposed to partner violence? A gender comparison did not find any differences.

Detainees who had any involvement in partner violence in the previous 12 months were, on average, only slightly younger (29 years old) than detainees with no involvement in recent partner violence (31 years old). While there were no significant differences between the groups on the basis of educational level attained, there were some differences in how income was generated during the past 30 days. Detainees with recent involvement in partner violence were significantly less likely to obtain an income from full-time work and were significantly more likely to obtain an income from family/friends or by engaging in drug dealing than detainees with no recent involvement in partner violence.

It is noteworthy that a significantly higher percentage of detainees involved in partner violence were taking care of dependent children (44%), than detainees with no partner violence involvement (34%).

Multivariate analyses in the form of a logistic regression were performed to determine which factors were risk markers for involvement in intimate partner violence. The factors explored were whether the detainee had been arrested or in prison in the past 12 months, was dependent on alcohol or drugs, tested positive to a particular drug, was currently being detained for a violent offence, experienced physical abuse as a child or witnessed partner violence as a child, or had dependent children at home.

Five of the factors examined were found to be significant risk markers for involvement in intimate partner violence (see Table4). Prior arrest, drug and alcohol dependency, having dependent children at home and experiencing child abuse were all significant risk markers for involvement in intimate partner violence.

Detainees who had been arrested in the past 12 months were 1.8 times more likely to be involved in intimate partner violence (Table4).

Detainees who were classified as drug dependent were twice as likely as non-drug dependent detainees to be involved in intimate partner violence. None of the individual drugs were found to be significantly associated with intimate partner violence, and this suggests that the level of illicit drug use (dependency), rather than the type of illicit drug is a greater predictor of involvement in intimate partner violence.

Alcohol dependency was also found to be a significant risk marker for involvement in partner violence.

Detainees with dependent children at home were two and a half times more likely to be involved in partner violence (Table4).

Police detainees who experienced physical abuse as a child were almost twice as likely as detainees who were not victimised as a child to report being involved in partner violence.

Witnessing partner violence as a child was not found to be a strong indicator of involvement in partner violence as an adult.

Table 4 : Risk markers for involvement a in partner violence b
Factorscoefficient Standard error Odds ratio
* statistically significant at p
a Separate models were run for perpetrators only and victims only and the results were very similar
b Some variables were not part of DUMA the first time the addendum was conducted. Sample only includes detainees from the third quarter of 2005
Source: AIC, DUMA collection 2005 [computer file]
Male -0.43 0.26 0.96
Dependent children at home 0.88* 0.19 2.41
Prior arrest (past 12 months) 0.58* 0.20 1.78
Prior imprisonment (past 12 months) -0.13 0.25 0.87
Alcohol dependent 0.42* 0.21 1.52
Drug dependent 0.79* 0.22 2.19
Experienced child abuse (physical) 0.62* 0.20 1.86
Witnessed partner violence as a child 0.34 0.20 1.41
Violent charge 0.29 0.27 1.34
Test positive to cannabis -0.06 0.20 0.94
Test positive to heroin -0.34 0.27 0.71
Test positive to methylamphetamine 0.12 0.22 1.13
Test positive to benzodiazepines -0.31 0.23 0.73
Constant -2.03*
Model 2 96.92*
df 13
(n) (631)

Concluding discussion

This analysis revealed a number of factors which appear to be associated with detainee involvement in partner violence. The bivariate analyses revealed that female detainees generally experienced more partner violence in their lifetime than males. The same pattern was apparent in relation to victimisation in the past 12 months, with female detainees significantly more likely to have been victims of recent partner violence.

However, it was also found that female detainees were more likely to be perpetrators of partner violence than male detainees. A possible reason for this finding is women who participate in criminal activity are more likely to have a history of illicit drug use and have experienced higher levels of abuse, economic hardship and other adversity in their lives than other women and men (Willis & Rushforth 2003). It is therefore not surprising that females in the DUMA sample were just as likely to have experienced intimate partner violence as they were to have perpetrated it.

Other possible explanations for these results relate to the use of self-report data. It is possible that the female detainees were more likely to reliably report perpetrating intimate partner violence than the male detainees. This is also a possible explanation for the lower rates of male victimisation, as reported by the male detainees.

When the risk markers for involvement in partner violence were examined, the results from the multivariate analysis indicated that gender of the detainees was not a significant predictor of involvement in intimate partner violence. The following factors were found to be significant risk markers for involvement in intimate partner violence:

  • arrest in the past 12 months
  • dependency on drugs (rather than the type of drug used)
  • dependency on alcohol
  • having dependent children at home
  • experience of physical abuse as a child.

Other Australian research found that women who experienced abuse during childhood were one and a half times more likely to experience any violence in adulthood (78% versus 49%; Mouzos & Makkai 2004). The research found that prior childhood victimisation was one of the significant risk factors for intimate partner physical violence.

Similarly, the experience of child physical abuse was found to be more prevalent amongst males and females who had experienced partner violence over their lifetime than males and females who did not experience any partner violence (ABS 2006). The results from this sample of police detainees tend to support the 'cycle of violence' premise.

Despite the plethora of research suggesting a link between witnessing partner violence as a child and subsequent involvement in partner violence as an adult, DUMA results indicate that witnessing partner violence as a child was not a strong predictor of involvement in partner violence as an adult.

Implications for policy

The findings outlined in this paper have a number of implications for policy development. In finding that gender was not a factor in involvement in partner violence for police detainees, it suggests that the development of intervention or prevention policies for police detainees needs to view intimate partner violence as not necessarily involving female victims and male perpetrators, but rather couples who engage in violent acts towards each other.

The finding that having dependent children at home increased the risk of detainee involvement in partner violence is significant because it not only has implications for those engaged in the partner violence, but also the children who may be witnessing the violence. While witnessing violence as a child was not a significant risk marker in the current study, the fact that much prior research has found a link (for partner violence in general) cannot be discounted (see Indermaur 2001).

In conclusion, the data showed that a large percentage of the detainees coming into contact with police were involved in intimate partner violence. Associated risk markers were drug and alcohol dependency, prior offending, and intergenerational experiences of violence. This suggests that there are differing levels of intervention required to address the issues of violence and drug use for persons who come into contact with the criminal justice system.

Within the family, early intervention could focus on identifying those at risk of abusing their children and partners and providing them with adequate support and positive alternatives to violence to enable them to cope with the stresses associated with parenting and relationships. Early signs of drug and alcohol dependency should also be addressed.

Outside the family, early intervention could focus on diverting first offenders into suitable programs to address the possibility of their reoffending, and becoming regular offenders. Previous research from the AIC's Drug Use Careers of Offenders project found that there is a one to two year window of opportunity to intervene and prevent the uptake of illicit drugs and the early onset of regular use following first offence. It noted that drug use initiation increased the risk of escalation to regular offending (Makkai & Payne 2003).

Acknowledgements

Professor Paul Mazerolle, formerly of the Queensland Crime and Misconduct Commission is acknowledged for developing the addendum in this study.

References

All URLs were correct at April 2007

  • Australian Bureau of Statistics (ABS). 2006. Personal safety survey Australia. ABS cat. no. 4906.0. Canberra: ABS
  • Cook PW (1997). Abused men: The hidden side of domestic violence. Westport, CN: Praeger Publishers
  • Fergusson DM, Horwood LJ & Ridder EM 2005. Rejoinder. Journal of marriage and family 67: 1,131-1,136
  • Indermaur D 2001. Young Australians and domestic violence. Trends & issues in crime and criminal justice no. 195.
  • Johnson H 2004. Drugs and crime: a study of incarcerated female offenders. Research and public policy series no. 63. Canberra: Australian Institute of Criminology
  • Johnson MP 2005. Domestic violence: it’s not about gender – or is it? Journal of marriage and family 67: 1,126–1,130
  • Makkai T 1999. Drug use monitoring in Australia: a brief description. Research and public policy series no. 21. Canberra: Australian Institute of Criminology
  • Makkai T & Payne J 2003. Drugs and crime: a study of incarcerated male offenders. Research and public policy series no. 52. Canberra: Australian Institute of Criminology.
  • McGregor K & Makkai T 2003. Self-reported drug use: how prevalent is under-reporting? Trends & issues in crime and criminal justice no. 260.
  • Mouzos J & Houliaras T 2006. Homicide in Australia: 2004–05 National Homicide Monitoring Program (NHMP) annual report. Research and public policy series no. 72. Canberra: Australian Institute of Criminology
  • Mouzos J & Makkai T 2004. Women’s experiences of male violence: findings from the Australian component of the international violence against women survey (IVAWS). Research and public policy series no. 56. Canberra: Australian Institute of Criminology. http://www.aic.gov.au/publications/ rpp/56/index.html
  • Skogan W 1986. Methodological issues in the study of victimisation, in Fattah E (ed), From crime policy to victim policy: reorienting the justice system. London: Macmillan
  • Statistics Canada 2003. Family violence in Canada: a statistical profile. Ottawa: Canadian Centre for Justice Statistics
  • Willis K & Rushforth C 2003. The female criminal: an overview of women’s drug use and offending behaviour. Trends & issues in crime and criminal justice no. 264. http://www.aic.gov.au/publications/ tandi2/tandi264.html

Dr Jenny Mouzos is a senior research analyst and Manager of the Crime Monitoring Program, and Lance Smith is a research assistant at the AIC.