After 20 years of domestic violence research, scientists can't avoid hard facts

by Nancy Updike
May/June 1999 

A surprising fact has turned up in the grimly familiar world of domestic violence: Women report using violence in their relationships more often than men.  This is not a crack by some antifeminist cad; the information will soon be published by the Justice Department in a report summarizing the results of in-depth, face-to-face interviews with a representative sample of 860 men and women whom researchers have been following since birth. Conducted in New Zealand by Terrie Moffitt, a University of  Wisconsin psychology professor, the study supports data published in 1980 indicating that wives hit their husbands at least as often as husbands hit their wives.

This from Dr. Felicity Goodyear-Smith, giving more detail on the study. 

Mark 
________________________________________________ 

This is part of the Dunedin Multidisciplinary Health and Development Study. 

 A summary of this study can be found in my  paper Goodyear- Smith FA; Laidlaw TM (1999).  "Aggressive acts  and assaults in intimate relationships: towards an understanding of  the literature", Behavioral Sciences & the Law (special issue on  'Threat Assessment and Management'), 17 (3), in press.

"The difference between 'assault' and 'aggressive acts' prevalence is clearly demonstrated by two different studies performed in New Zealand on the same birth cohort.  The Dunedin Multidisciplinary Health and Development Study (DMHDS) is an extensive longitudinal cohort study of the health, development and behaviour of 1037 New Zealanders born in Dunedin between 1 April 1972 and 31 March 1973.  When this cohort was 21 years of age, studies were conducted about their perpetration and victimisation rates of partner violence, and their overall experience of assault over the proceeding 12 months.  Both these interviews were part of a battery of assessments of 941 members of the cohort conducted over an entire day.  The partner violence questions were embedded in a 50 minute standardised interview about intimate relationships in the morning, and the assault interview was a 22 minute interview in the afternoon. 

The assessment of both perpetration and victimisation of partner violence for this cohort used the Conflict Tactics Scales (CTS) to ask about aggressive acts being performed (Magdol et al., 1997).  Respondents answered twice: first about their behaviour towards their partner, and second, about their partner's behaviour towards them.  Women reported perpetrating more partner violence than men, and men reported more victimisation than women.  When subjects not involved with a partner in the proceeding year were excluded, it was found that nearly all the women (95%) and a large majority of the men (86%) reported having performed an act of verbal aggression against a partner.  Similarly, prevalence rates of perpetration of physical violence by women were significantly higher than those for men (37% compared to 22%).  This included rates for severe physical abuse (such as kicking, hitting, biting, hit with a weapon, use or threat of use of a knife or gun) which 19% of women and 6% of men reported that they had used on their partner. 

Analysis of victimisation rates gave confirmatory findings.  More men (34%) than women (27%) reported being a victim of physical violence perpetrated by their partner.  This study also found that while women from all social strata were liable to be violent, there was an increased risk for men to be violent if they were poorly educated, unemployed, and lacked social supports.  This data suggests that men from higher socio-economic groups with better educational status are less likely to engage in violent acts against women. 

In contrast, the study looking at the rates of physical assault in  the preceding 12 months for this cohort gave different results, with domestic assault rates much lower than the violent behaviour rates  reported above (Martin et al., 1998; Langley et al., 1997).  In a semi-structured face-to-face interview, subjects were asked if anyone had  deliberately harmed, or attempted or threatened to deliberately harm them  in the preceding 12 months.  If the answer was 'yes', details were then  elicited including the frequency and nature of the attacks, and the gender and relationship of the assailant. 

Nearly half of the 482 men and a quarter of the 462 women reported at least one physical assault, attempt or threat in the previous  12 months.  While men were the predominant offenders, 15% of the  assaults on men, and 25% of the assaults on women, were perpetrated by  women.  The women who reported assault in the past 12 months were  more likely to be currently living without a partner, either alone or with children and living on the Domestic Purposes Benefit. 

While many of the subjects owned to various acts of physical  inter-partner violence given or received, far fewer of these defined these events as "assaults". Only 3% of the men and 11% of the women reported assaults by partners. These percentages cannot be directly compared with the victimisation rates from the other study, because the samples are somewhat different.  The assault study includes subjects who were not in any relationship, and more men had not been in a relationship than women in the overall cohort.  When the actual numbers of men and women for each study are calculated, however, it is found that 144 men report being subjected to violent acts by female partners, but only 14 of those classify these acts as "assaults" or view them as "deliberately intended to harm".  In contrast, 118 women report being subjected to violent acts by their male partners, but many more of them (51) consider that these acts are "assaults". Women reported significantly more injuries for which they sought medical treatment than men. 

A primary difference between the two studies are the definitions  used.  It seems that women are more likely to define slaps, hits, and  punches from their partner as "assaults", while men are less likely to do so. 

A further study was conducted on 360 members of the cohort  and their intimate partners (Moffitt, 1987).  It was found that perpetrators and victims generally agreed about the extent to which perpetrators engaged in physically violent and psychologically abusive behaviours, and that this agreement did not vary with the perpetrator's gender.  The level of agreement in relation to specific acts was only modest, but when responses were aggregated, the correlation between victim and perpetrator reports was very strong. 
 

Magdol L, Moffitt T, Caspi A, Newman D, Fagan J & Silva P (1997).  "Gender differences in partner violence in a birth cohort of 21-year-olds: bridging the gap between clinical and epidemiological  approaches" Journal of Consulting & Clinical Psychology, 65  (1), 68-78. 
 

Moffitt TE & Caspi A (1998). Annotation: implications of violence between intimate partners for child psychologists and  psychiatrists, Journal of Child Psychology & Psychiatry, 39 (2),  137-144. 
 

Moffitt TE, Caspi A, Silva PA(Nov 1996). Findings about partner violence from the Dunedin Multidisciplinary Health and Development Study, University of Otago Medical School, Dunedin. 
 

Dr Felicity Goodyear-Smith 
Research Fellow 
Department of Psychiatry & Behavioural Science 
University of Auckland 
PB 92 019 Auckland 
New Zealand 
tel 64 9 373 7599 Ext 2357,; 021 897 244. 
___________________________________________