The main purpose of collecting data on domestic violence is that by sharing information between criminal justice agencies it ensures that these agencies are kept up to date and offenders are held accountable for their actions. (Hall & Wright (2003) p3). The main collection of domestic violence statistics can be found in government surveys such as the British Crime Survey and the Computer- Assisted Self-Interviewing Component (CASI). The British Crime Survey, first conducted in 1982, provides estimates, produces trends and determines the risks for certain areas of the country. The computer-assisted method (CASI) is a questionnaire that respondents can complete themselves, the questions are predominantly ‘yes’ or ‘no’ questions ensuring higher levels of confidentiality. According to the 2001 British Crime Survey there were 499,000 cases of domestic violence in 2000 (Kershaw (2001) p28), the Self-Completion questionnaire found that 4.2% women and men had been physically assaulted by a current or former partner in the last year and 5.9% of women and 4.9% of women experienced frightening threats, these percentages make up the estimated 6.6 million incidents and two deaths a week that occur as a result of domestic violence in 1995 (Mirrlees-Black (1999) pvii).
In addition to the government funded national surveys, more local surveys are also conducted adding more detail to the larger research findings. However, the lack of “sound methodological grounds” (Mirrlees-Black (1995) p4) found in local research means the findings lack ecological validity. Independent research surveys into domestic violence are also conducted, a main example is the North London Domestic Violence Survey conducted by Mooney in 1993 in which 1000 women took part in in-depth interviews about their experiences. The results of this survey estimated that 12% of women had experiences actual physical violence from a partner or ex-partner in the last year, 8% were injured and 6% were raped (Mooney (2000a) p163). Other local surveys include, Russell (1985), Hall (1985), Hanmer & Saunders (1984), and Painter (1991) (Walby (1999) p2)
The variation found between government and independent surveys is immediately obvious when comparing the rates of domestic violence found. An example of this is in comparing the results of the BCS, CASI and the North London Domestic Violence Survey:
The obvious conclusion to be drawn from this comparison is that the huge differences shown between the national surveys and the North London survey are as a result of under-estimates in the first two examples and possibly also in the third.
The British Crime Survey has been criticised for the small sample size on which the national estimates for domestic violence are based (Kershaw (2001) p29), this results in a lack of ecological validity as the results cannot be used to improve national services provided for victims of domestic violence. The BCS statistics are taken from face to face interviews, this may not be the most effective method as respondents may not wish to disclose such private information to a stranger, the amount they are willing to disclose may also change over time, the presence a partner at the time of the interview may also inhibit reporting of domestic violence to the survey (Mirrlees-Black (1999) p70). The BCS also excludes those who have been moved to emergency shelters and those that are homeless, according to Walby, “this methodological issue can have major implications for the theoretical understanding [of domestic violence] if the most abused and most recently abused group are underrepresented” (Walby (1999) p3). It was found that samples based on women outside of the home, for example those in shelters shows a much higher rate of abuse from national surveys (Walby (1999) p3). This is an important point as by excluding abused women from the surveys domestic violence will not be tackled effectively as the services provided by the government are the result of limited statistics. The household surveys themselves could be expanded; Walby suggested that within each household the socio-economic status of each member should be calculated, this would help to investigate intra-household power and violence more effectively; also if the government is aware of the economic and social status of the victims they would be able to provide more effective services.
The BCS is published annually, this method has been criticised as when victims are asked about their experiences over the year less serious incidents are less likely to be remembered, therefore it would appear that there is an increase in more serious assaults as these incidents are more easily recalled (Mirrlees-Black (1999) P17). This would result in the false assumption that there is a rise in more serious domestic violence, in turn leading to ineffective changes in services available as the statistics are not reliable.
The “life-time measure” has been found to be a more reliable method of questioning (Mirrlees-Black (1999) p27) as it can start to assess the long-term consequences of domestic violence. The life-time measure tackles the problem of household surveys to an extent as it includes abuse suffered over a longer period of time. Along with the danger of underestimates arising from crime surveys there is also the possibility of overestimates, this could be the result of respondents exaggerating their experiences, there is no way to assess this it is simply a factor to be aware of when interpreting domestic violence statistics. The inclusion of incidents that do not meet the legal definition of a crime could also result in overestimates; this links back to the need for a universal definition of domestic violence that is used by all crime surveys.
In 1996, as a result of the government addressing some of the weaknesses found in the original British Crime Survey the quantitative Computer-Assisted Self-Interviewing component was combined with the more qualitative interview technique, this, according to Walby resulted in the 1996 BCS being classed as a second generation survey, prior to this is was defined as a more generic survey (Walby (1999) p1). The aim of this was to provide a more reliable estimate of the prevalence of domestic violence. Despite this progress, there were some criticisms of the computer-assisted method of interviewing victims; a main example is the layout of the questionnaire, it only allows a yes or no answer for most of the questions, this results in less detail being collected about the victim, the incident, and the attacker so although more results are collected due to the focus on confidentiality and structure of the questions, it does not provide the same amount of detail as the face to face interviews do.
According to Felson in his article ‘Reasons for reporting and not reporting domestic violence to the police’, there are three main incentives and four main cost factors for women when reporting domestic violence. These incentives and cost factors can be addressed when looking at the methodology of victim surveys in order to make these surveys more effective. The four main cost factors are; the desire to protect the offender from criminal prosecution (Pagelow 1984), the fear of reprisals (Gottfredson & Gottfredson 1987), if the victim is engaged in illegal activities e.g. drugs (Black 1983) and finally, opportunity costs, e.g. legal process. These cost factors can be tackled through not only the methodology used in the crime surveys, for example ensuring maximum confidentiality, but they can also be addressed through educating not only the victims themselves but also the authorities and the general public so that it becomes generally accepted that domestic violence is not acceptable and help is available to the victims regardless of their circumstances.
The three main incentives are; protection (Gottfredson & Gottfredson 1987), justice (Black 1983), and a civic duty to protect others (Gottfredson & Gottfredson 1987; Smith & Maness 1976). In order for victims of domestic violence to feel confident about divulging their experiences to the police or victim surveys they need to feel confident that these three incentives are met.
The Domestic Violence, Crime and Victims Act passed in 2004 is a step forward for the government dealing with domestic violence, it has been described widely as “the biggest overhaul of domestic violence legislation in 30 years” (). The Act aims to put victims at the centre of the criminal justice system. This legislation, along with the various campaigns launched by the government, e.g. the ‘zero tolerance’ and BBC ‘hitting home’ campaigns, have started to make the public more aware of domestic violence and the extent of the problem. This legislation helps to maintain the incentives described by Felson. As a result in this increasing focus from the government and the general public domestic violence will become more accepted in society as being a criminal act, leading to women feeling more confident that their case will be dealt with effectively and that they will receive the support they need, this will in turn result in more research as higher levels of domestic violence are reported.
To conclude, crime victim surveys have proved to be an “indispensable tool” (Walby (1999) p2) for analysing domestic violence despite the hesitations and criticisms of some researchers. However, there is always room for improvement in the methodology used in crime victim surveys, examples discussed include the lack of confidentiality found in the British Crime Survey, the isolation of victims outside of the household such as those moved to emergency accommodation and the lack of detail collected in the computer-assisted method. The definition of domestic violence is a factor that continues to limit the reliability of the crime statistics presented; a universal definition is required, particularly between the agencies working with the victims of domestic violence. Despite the limitations found in the research surrounding domestic violence there have been some major progressions, the Domestic Violence, Crime and Victims Act passed in 2004 along with the campaigns launched by the government and the BBC to promote awareness of domestic violence have helped to educate the authorities and the general public on the extent of domestic violence and how it can be dealt with.
References:
-
Ellison, L. (2002), ‘Prosecuting Domestic Violence without Victim Participation’, Modern Law Review, Vol. 65, No.6. pp.834-858.
-
Kershaw, C. et al (2001), The 2001 British Crime Survey, Home Office Statistical Bulletin 18/01, London, Home Office. pp28-99
-
Maguire, M. (2002), Crime Statistics: The ‘Data Explosion’ and its Implications in Maguire, M. et al (eds.), The Oxford Handbook Of Criminology. pp322-375
-
McLaughlin, E. & Muncie, J. (2001) The Sage Dictionary of Criminology. pp117-118
-
Mirrlees-Black, C. (1999), Domestic Violence: Findings from a new British Crime Survey Self-Completion Questionnaire, HORS 191, London, Home Office. pp1-73
-
Mooney, J. (2000a), Gender, Violence and the Social Order. pp141-217
-
Walby, S. (1999), Comparing Methodologies used to study violence against women. pp1-10
Sources Used:
-
Anderson, K. L. & Umberson, D. (2001) ‘Gendering Violence’, Gender and Society, Vol 15, No.3, pp358-380
-
British Crime Survey Self-Completion Questionnaire, Home Office Research and Statistics Directorate Research Findings No.86 available at
-
Caralis, P.V. & Musialowski, R.(1997), ‘Women’s experiences with Domestic Violence and their attitudes and expectations regarding medical care of abuse victims’, Southern Medical Journal, Vol 90, No.11. pp1075-1083.
-
Ellison, L. (2002), ‘Prosecuting Domestic Violence without Victim Participation’, Modern Law Review, Vol. 65, No.6. pp.834-858.
-
Felson, R. et al. (2002), ‘Reasons for reporting and not reporting domestic violence to the police’, Criminology, August 2002, Vol. 40, No.3, pp.617-647
-
Felson, R. & Messner, S. (2000), ‘The Control Motive in Intimate Partner Violence’, Social Psychology Quarterly, Vol. 63, No.3, pp86-94
-
Fulton, D. R. (2000), ‘Recognition and Documentation of Domestic Violence in the Clinical Setting’, Critical Care Nursing Quarterly, Vol 25, No.6. pp26-34.
-
Home Office Website: .
-
Hoyle, C. (2000), Police response to Domestic Violence, British Journal of Criminology, Vol. 40 , pp.14-26.
-
Kershaw, C. et al (2001), The 2001 British Crime Survey, Home Office Statistical Bulletin 18/01, London, Home Office.
-
Maguire, M. (2002), Crime Statistics: The ‘Data Explosion’ and its Implications in Maguire, M. et al (eds.), The Oxford Handbook Of Criminology, Oxford, Oxford University Press.
-
McLaughlin, E. & Muncie, J. (2001) The Sage Dictionary of Criminology, London, Sage.
-
Mirrlees-Black, C. (1999), Domestic Violence: Findings from a new British Crime Survey Self-Completion Questionnaire, HORS 191, London, Home Office. Also available at
-
Mooney, J. (2000a), Gender, Violence and the Social Order, Basingstoke, Macmillan Press.
-
Walby, S. (1999), Comparing Methodologies used to study violence against women available at
- http://www.womenandequalityunit.gov.uk/domestic_violence/#dv_crime_victims_bill