Australian Institute of Criminology

Skip to content

Studying victims of crime

Accessing reliable and appropriate samples of victims for analysis is difficult. The three most common methods by which information about victims is gathered are through crime statistics recorded by police, victimisation surveys or victimisation studies. The strengths and weaknesses of each of these approaches are discussed below.

The risk of re-traumatising or ‘secondary victimisation’ is a significant barrier to in-depth research with victims. The concept of secondary victimisation should not to be confused with the issue of secondary victims though the two terms are often closely related. Secondary victimisation (also referred to as ‘second assault’ or ‘second rape’) refers to instances where a victim is further victimised or traumatised through negative experiences during the criminal justice process and/or by support organisations (Campbell & Raja 1999; Campbell et al. 2001). Montada (1994) notes that this is experienced by victims as ‘further violation of legitimate rights or entitlements’ (Orth 2002: 314). Secondary victimisation can also be used to describe individuals who become traumatised through contact with the primary victim, such as family, friends or support workers. Throughout this report it is these individuals who are referred to as ‘secondary victims’.

The likelihood of a victim of crime being re-traumatised by their involvement in the justice system has been the focus of research for a number of years, with a particular focus on victims of sexual assault (see Campbell & Raja 1999; Koss 2000; Campbell et al. 2001). Orth (2002) studied the impact of participation in criminal proceedings by surveying a sample of 137 victims of violent crime in Germany. Respondents reported what they perceived the impact of their involvement with the criminal justice system to be—measured from very negative to very positive. Other measures included satisfaction with the outcome, severity of the punishment, justice and level of stress. Orth found that victims were at risk of secondary traumatisation through criminal proceedings. In particular, participation had negative impacts of the victim’s ‘faith in a just world’ and trust in the legal system (Orth 2002: 321). Other studies have also highlighted the potentially traumatising nature of interactions with criminal justice and/or support agencies post crime. Of 286 mental health professionals surveyed by Campbell and Raja (1999), 81 percent felt that participation in the legal system was harmful to victims of sexual assault. Further, 84 percent believed that the responses of community professionals could also contribute to secondary victimisation. Ways in which this occurred included undergoing a forensic medical exam and harmful counselling practices (Campbell & Raja 1999).

Herman (2003) outlines a number of ways court processes can impact on the victim; in particular, the negative psychological effect of having to recount their experiences in an open court, sometimes in front of the perpetrator. Further, the adversarial nature of the trial may expose the victim to attacks on their credibility, leaving them feeling traumatised and invalidated (Herman 2003). Despite many criminal justice agencies and court systems making important changes to their processes in an effort to better support victims of crime (see Outtrim 1999; McGregor, Renshaw & Andrevski 2003 for some examples), re-traumatisation and secondary victimisation remain a key concern. Currently there are no studies available which examine the likelihood of secondary victimisation through participation in research. However, it is still an important concern, particularly if the research requires the participant to recount their experiences. The National Statement of Ethical Conduct in Human Research (2007), hereafter referred to as the National Statement (2007), has clear directives on the harm to participants posed by research. The following harms possibly associated with conducting research are listed in the National Statement (2007):

  • ‘psychological harms: including feelings of worthlessness, distress, guilt, anger or fear-related, for example, the disclosure of sensitive or embarrassing information…’;
  • ‘social harms: including damage to social networks or relationships with others…social stigmatisation…’ (National Statement 2007: 16).

It is not difficult to see how research that involves victims could potentially harm the participant. This risk is likely to increase the more the participant is required to discuss and relive their experiences. To minimise the harm to participants, human research ethics committees often require strict guidelines around accessing, interviewing/surveying victims and analysing and reporting on the data, and this may limit the scope of potential research. As a result, victim research can be very difficult to conduct and researchers therefore look for alternative ways to collect information associated with this population. The most common ways are through police recorded crime statistics, victimisation surveys or self-report interviews.

Recorded crime statistics

In Australia and overseas, crime statistics recorded by police provide one of the most important and reliable sources of information regarding crime. In Australia, two types of police statistics are available, those collected by each state and territory police force, or the national statistics collected by the ABS in its publication Recorded crime, victims, Australia. The data in these types of publications provide valuable insights into the volume of crime in Australia and associated temporal patterns. The ABS also disaggregates the information by certain types of crime as well as providing further data on the age and demographics of victims and the basic circumstances of the crime such as where it occurred and weapon used (ABS 2012). Researchers have used police recorded crime statistics for a wide variety of purposes including examining fluctuations in crime and associated characteristics (Bricknell 2008; AIC 2014), local area analysis (Donnelly & Snowball 2006) and to determine the impact of changes to policing methods, policy or crime prevention (Chilvers, Korbelnikoff & Ramsay 2002).

The key criticism of police crime statistics is that they reflect only the volume of reported crime. Not all crime is reported to police, this varies depending on the crime. Traditionally, property crime has had higher rates of reporting than violent crime, particularly sexual assault (AIC 2013). Reasons given for not reporting an incident to police include a belief that the incident was minor/trivial, a lack of confidence in police and a decision to deal with it personally (ABS 2012). Police crime statistics do not include information pertaining to those victims who do not report their crime; these may form a sub-population who could be qualitatively and quantitatively different to those who do. Further, recorded crime statistics are constrained by what is considered, at that time, to be an offence. MacDonald (2002) included victimless crimes (such as illegal drug use) as well as those not reported to police in the ‘dark figure’ of crime that is missed in official recorded statistics. However, police statistics remain a valuable data source. As noted above, they are regularly used for a wide variety of reasons and as long as there are no significant fluctuations in the rate of reporting across time or location, police statistics will reflect any changes in actual offending (Weatherburn 2011). Yet, for researchers seeking to gain a greater understanding of victimisation beyond volume and basic characteristics, recorded crime statistics may provide limited assistance.

Crime victimisation surveys

Crime victimisation surveys, such as the ABS’ Crime Victimisation Survey (CVS) or the International Crime Victimisation Survey (ICVS) attempt to capture information related to individuals who do not report their crime to police (as well as those that have reported). This helps to better record the prevalence of a criminal offence in the community. Methods vary across surveys, however the ICVS interviews between 1,000 and 2,000 individuals drawn from national or urban populations using computer-assisted telephone interviewing (CATI) and face-to-face interviewing (van Kesteren, van Dikj & Mayhew 2013). Standard questions focus on personal and household experiences of crime, as well as addenda added each year. Addenda topics have examined issues such as the support received by victims, fear of crime, and police response interviewing (van Kesteren, van Dikj & Mayhew 2013). The ABS survey is similar in its method, though on a smaller national scale. Compared with the ABS publication, Recorded crime—victims, Australia, the CVS estimates numbers of individuals who did not report. Further, it surveys why they did not report as well as other characteristics such as their employment and relationship to offender (ABS 2013).

Weatherburn (2011) identifies a number of weaknesses with crime victimisation surveys, specifically, that they:

  • are limited in their ability to collect information on ‘victimless’ crimes such as illegal drug use or rare crimes such as homicide; and
  • they can compare poorly to police recorded crime statistics in terms of the detail available about the incident of crime (Weatherburn 2011).

Further, like police statistics, crime victimisation surveys are epidemiological in nature. They reduce victimisation to quantitative information that allows researchers to examine the trends and patterns. The inclusion of individuals who do not report to police may give researchers a better indication of the prevalence of victimisation both within Australia and internationally. However, like police recorded crime statistics, victimisation surveys are limited in their ability to provide detailed information on the impact and experiences of victims of crime.

Victimisation studies

Given the limitations of both police statistics and crime victimisation surveys, the most common method for collecting detailed information about the experiences of victims is via direct contact. Victimisation studies use a variety of methods including semi-structured interviews, self-report questionnaires, or the application of psychometric tests. The benefit of these approaches is that they provide the detail currently missing from recorded crime statistics and victimisation surveys. For example, Field, Zander and Hall (2013) recently used semi-structured interviews to examine the views of 11 victims of crime regarding forgiveness. Participants were recruited via ads in local newspapers throughout Western Australia and the results were analysed using qualitative coding methods. They found that victims of crime viewed forgiveness and the ability to forgive as closely connected to their own concepts of psychological wellbeing. This contrasted with models presented in other literature that focused on the interpersonal and social outcomes of forgiveness (Field, Zander and Hall 2013). Field, Zander and Hall’s research is an example of how these types of studies have the ability to more closely describe the true nature and impact of crime on victims.

However, this approach has limits, in particular, the trade-off between detail and generalisation. Field, Zander and Hall’s (2013) study interviewed only 11 participants, which limits the ability to generalise findings to the wider population of victims. Further, studies that explore victimisation in this depth are not often comparable. For example, the differences in sample, methodologies and types of analysis used will mean that two studies looking at the experiences of sexual assault victims cannot be compared. Finally, re-traumatisation is a key concern with this type of research, as it often requires the victim to recount or remember in detail the experience of the crime. As stated previously, efforts to mitigate this risk often result in limitations on the way victims can be sampled, questioned and their data analysed.

Overall, researchers have traditionally used a variety of methods to study victimisation both within Australia and internationally. Police-recorded crime statistics and victimisation surveys are valuable sources of information on the volume and trends in crime but lack the detail needed to fully explore the impact of violence on the victim. In contrast, studies of victimisation can obtain that level of detail, but often have a narrow focus, which limits their ability to be applied to the wider population.