Trends & issues in crime and criminal justice

No. 298: Crime victimisation in Australia : key findings of the 2004 International Crime Victimisation Survey

Holly Johnson
ISBN 0 642 53879 4 ; ISSN 0817-8542
April 2005

Most Western industrialised countries, and many developing countries, have come to rely on victimisation surveys to provide estimates of:

• rates of victimisation;
• variations in rates of victimisation among segments of the population;
• the percentage of crimes reported to police;
• reasons for not reporting;
• fear of crime and perceptions of safety; and
• efforts undertaken by individuals to protect themselves from crime.

Victimisation survey data provide an important complement to official statistical data produced by police and other criminal justice agencies as they include both crimes reported to the police and crimes not reported. The Australian Bureau of Statistics conducts a crime victimisation survey, the Crime and safety survey, on a regular basis, most recently in 2002 (ABS 2003). The ICVS is broader in scope and can be adapted to focus on issues of specific policy interest to the Australian Government. The ICVS is an international project involving approximately 60 countries worldwide and, as a result, the Australian findings can be compared with those of other countries for rates of victimisation and perceptions of crime and the criminal justice system (see van Kesteren, Mayhew & Nieuwbeerta 2000). The United Nations Office on Drugs and Crime will be undertaking comparisons across countries participating in the 2004 ICVS.

The 2004 International Crime Victimisation Survey

Funding for the Australian component of the ICVS was provided in two parts: the Australian Government Attorney-General's Department funded a random national sample of 6,000 respondents; and the Australian Government Department of Immigration and Multicultural and Indigenous Affairs funded a booster sample of 1,000 migrants. Additional socio-demographic questions were added to identify place of birth and language spoken at home by migrants. The sample of 7,000 is considerably larger than past Australian samples of the ICVS, and typical samples in other countries, which have been in the order of 2,000. The enhanced sample in the 2004 Australian ICVS is designed to enable more detailed analysis than has been possible in the past.

The in-scope population for the 2004 ICVS was all adults 16 years of age and older who were residents of private households in Australia. Households were selected using random digit dialling and data were collected by computer-assisted telephone interviewing (CATI). A response rate of 53 per cent was achieved. The sample was weighted to accurately represent the age, gender and birthplace of the Australian population according to the 2001 census of population and housing.

Rates of victimisation

The ICVS asks respondents about their experiences of select types of crimes in the preceding five years. From this, five-year rates and one-year rates for 2003 can be calculated. Three personal and six household crimes were included in a standardised questionnaire. Personal crimes include assaults and threats, robbery and theft of personal property. Household crimes include burglary, attempted burglary, motor vehicle theft, theft from motor vehicles, motorcycle theft and bicycle theft (see Box 1).

Victimisation rates are calculated per 100 persons and per 100 households. These are prevalence rates which count the percentage of people or households victimised once or more. Overall, half (52%) of the sample reported at least one experience of crime over the previous five-year period and 17 per cent were victimised during 2003. For personal crimes 29 per cent had been victims over the five-year period and nine per cent in 2003. Rates of personal crime were highest for assaults/threats and personal theft and lowest for robbery (Figure 1).

Rates were higher for household crimes than for personal crimes: 39 per cent of households reported at least one of the household crimes measured by this survey during the five-year period, and 11 per cent experienced a household crime in the 12 months prior to the survey. Theft from motor vehicles occurred most frequently, followed by burglary and attempted burglary. Theft of motor vehicles, motorcycles and bicycles were reported by less than 10 per cent of the sample over the five-year period and less than two per cent in the past year (Figure 2). These rates are based on all households. Restricting the analysis to just those households who were owners of motor vehicles, motorcycles and bicycles, five-year rates are as follows:

• motor vehicle theft: seven per cent of motor vehicle owners;
• theft from motor vehicles: 21 per cent of motor vehicle owners;
• motorcycle theft: five per cent of motorcycle owners; and
• bicycle theft: 10 per cent of bicycle owners.

Trends in victimisation rates

Comparison of the 2004 crime victimisation rates with the 2000 survey show statistically significant declines for both one- and five-year rates (Table 1). The percentage of persons who reported at least one incident of crime over the previous five-year period was 52 per cent in 2004, a statistically significant decline over 55 per cent in 2000. This decrease is largely due to the drop in personal theft which is one of the highest frequency crimes.

One-year rates indicate more about current crime conditions. The overall one-year victimisation rate dropped from 24 per cent to 17 per cent between the two surveys. Three of the nine crimes asked about showed significant declines. These were personal theft, burglary and theft from motor vehicles. These patterns mirror trends in police recorded crime in Australia which also show recent declines in rates of property theft and burglary (AIC 2004).

Table 1: Comparative rates of victimisation, 2000 and 2004

	One-year rates		Five-year rates
	2000	2004	2000	2004
Total victims	24	17*	55	52*
Assault/threats	6	5	19	18
Robbery	1	1	4	3
Personal theft	7	4*	18	14*
Burglary	4	3*	14	13
Attempted burglary	3	2	11	10
Motor vehicle theft	2	1	7	7
Theft from motor vehicle	7	5*	19	19
Motorcycle theft	0.1	0.1	1	1
Bicycle theft	2	1	7	6
* Difference is significant p<.05 Due to adjustments made to the 2000 data file to ensure compatibility with the 2004 survey, figures differ from those published in Carcach & Makkai (2003) Source: Australian Institute of Criminology, International crime victimisation survey, 2000 and 2004 [computer files]

Variations in victimisation rates

It is a common finding in crime victimisation research that risk of victimisation is not evenly distributed in the population: people and households within certain socio-demographic groups report higher rates. One-year rates of victimisation are used in this analysis as some personal characteristics, such as age, income and others, change over time and may not reflect the person's situation five years previously.

Bivariate analysis identified correlates of personal crime that are consistent with results of other victimisation surveys. These include age (under 25), marital status (unmarried), high residential mobility, unemployment, and frequently going out in the evenings for recreational purposes (ABS 2003; Mihorean et al. 2001). One finding that was inconsistent with prior research is the positive relationship between household income and personal crime: in the ICVS, those in the lowest income category (less than $400 per week) had the lowest rates of personal crime. The same correlates hold true for the violent offence of assault/threats. In addition, those who speak a language other than English at home report lower rates of assaults or threats.

A number of household-level variables are important correlates of household crime. For example, within the one-year period, rates were highest for higher income households and low for households that were stable for more than one year at the current postcode. Rates of household crime do not vary significantly according to type of dwelling, Indigenous status, or speaking a language other than English at home.

Predictors of victimisation

Many of the factors that are associated with higher rates of personal or household victimisation are inter-rated. For example, young people are more likely than older people to be single, to be looking for work, and to be active outside the home in the evenings. Using logistic regression, the most important risk factors of personal and household victimisation can be identified, while holding constant the effects of the others. For a more detailed explanation of the method, see the full report (Johnson 2005).

As shown in Table 2, five factors were significant risk factors of personal victimisation in the previous year while controlling for the effects of others:

• marital status: those who were single, separated or divorced, or living in a de facto relationship had higher odds of personal crime;
• income: persons in lower income households (under $400 per week) had reduced odds of personal crime;
• residential stability: persons who were living at their current postcode less than one year had significantly higher odds of personal crime;
• unemployment: unemployed persons had higher odds of personal crime; and
• night-time activities: those who participated in recreational activities outside the home almost every evening had heightened odds of personal victimisation.

Analysis of assault/threats separately from other personal crimes identified the same risk factors, with the addition of increased cent of the sample reported one crime type, 14 per cent reported two, and 10 per cent three or more (Figure 3). Twelve per cent of all households experienced more than one type of household crime while six per cent of all persons reported more than one personal crime.

Within crime types, many victims also reported multiple victimisations. Those who reported being victimised were asked how often the crime had occurred within the previous year (2003). Over this one-year period, 68 per cent of all victims reported experiencing one incident of crime, 19 per cent reported two and 13 per cent experienced three or more. The crime most likely to be repeated against the same victim was assault or threat: 19 per cent of assault/threat victims experienced three or more assaults or threats within the one-year period (Table 4). Only one factor - being male - differentiated repeat victims of personal crime from one-time victims. And none of the factors tested for, including income, time at postcode, language spoken at home or Indigenous status, were significant risk factors for repeat experiences of household crime.

Table 2: Risk factors for personal victimisation, logistic regression

	Adjusted odds ratios	SE	95% CI
Gender (male)	0.91	0.09	0.77-1.08
Age (younger)	1.12	0.12	0.89-1.41
Marital status (unmarried)	1.85**	0.10	1.52-2.25
Income (lower)	0.66**	0.15	0.49-0.89
Time at postcode (less than one year)	1.55**	0.12	1.22-1.97
Unemployed	1.79**	0.23	1.15-2.79
Evenings out almost every day	1.96**	0.12	1.55-2.49
Language other than English	0.96	0.12	0.76-1.2
Indigenous	1.22	0.24	0.76-1.96
-2 log likelihood 3959.6 Model chi square 150.6** (9 df)
* p<.1; ** p<.05 Source: Australian Institute of Criminology, International crime victimisation survey, 2004 [computer file]

Table 3: Risk factors for household victimisation, logistic regression

	Adjusted odds ratios	SE	95% CI
Income (lower)	0.61**	0.12	0.49-0.77
Time at postcode (less than one year)	1.32**	0.12	1.05-1.67
Language other than English	0.94	0.11	0.75-1.16
Indigenous	1.23	0.22	0.79-1.9
-2 log likelihood 4713.64 Model chi square 25.99** ( 4 df)
* p<.1; ** p<.05 Source: Australian Institute of Criminology, International crime victimisation survey, 2004 [computer file]

Perceptions of personal safety

Overall, the majority of Australians feel relatively safe in their local environment: 72 per cent reported feeling fairly or very safe while walking alone in their local area after dark (Figure 4). Fifty-eight per cent of respondents felt it was unlikely their homes would be broken into in the coming year while 29 per cent said it was likely. Seven per cent felt it was very likely. Comparison with the 2000 ICVS suggests perceptions of personal safety have improved recently: 64 per cent of Australians in 2000 felt safe while walking alone in their local area after dark (40 per cent felt fairly safe and 24 per cent felt very safe).

Women were more than twice as likely as men to say they feel a bit unsafe and four times as likely to say they feel very unsafe walking alone in the local area after dark. Improvements to feelings of safety since 2000 were applicable to both women and men. In 2004, 21 per cent of women felt very safe walking alone in their local area after dark, up from 13 per cent in 2000. The figures for men were 44 per cent in 2004 and 35 per cent four years previously. Feeling unsafe is also associated with being younger, speaking a language other than English at home, and being the victim of a crime (while controlling for time at postcode, income and Indigenous status).

Table 4: Number of victimisations in 2003 by crime type (per cent)

	One	Two	Three or more
Total	68	19	13
Personal crimes	72	15	13
Assault/threat	67	14	19
Robbery	77	16	7
Personal theft	86	11	3
Household crimes	74	18	8
Burglary	84	12	4
Attempted burglary	84	11	5
Motor vehicle theft	95	--	-
Theft from motor vehicle	83	12	5
Motorcycle theft	86	--	--
Bicycle theft	85	9	6
- less than one per cent -- fewer than five cases Source: Australian Institute of Criminology, International crime victimisation survey, 2004 [computer file]

Reporting crimes to the police

The crimes most likely to be reported to police in the 2004 ICVS were those involving substantial property losses that were likely to be insured, such as theft of motor vehicles and motorcycles (Figure 5). Burglary, which involves invasion of one's home and property loss or damage, and has the potential to increase fear, also had high rates of reporting (84%). Robbery, which involves both property loss and violence, was the personal crime most likely to be reported, at just over half of all incidents. The crime least likely to be reported to police was assault/threat (37%).

There was variation among victims who reported to police. For example, among assault/threat victims, those most likely to report to police were older, separated or divorced, living in relatively low income households, unemployed, Indigenous, confronted with a weapon, physically attacked as opposed to threatened, physically injured, and confronted by multiple offenders (Table 5). In logistic regression analysis, all of these indicators remained significant predictors of reporting assault/threats to police, with the exception of marital status, unemployment, and presence of a weapon.

Burglary victims who had higher reporting rates tended to be living in higher income households, living at their current postcode for a year or more, and had property loss as a result of the burglary, especially those who lost property valued at over $1,000 (Table 6). These remained significant risk factors in a logistic regression model.

Table 5: Rates of reporting assault/threat by characteristics of victims (per cent)

	Reported to police
Age
16-24	29*
25 and over	39
Marital status
Single	32*
Married	39
De facto	34
Divorced/separated	49
Widowed	--
Household income
<$400 per week	48*
$400 or more	36
Main activity
Working	38*
Looking for work	54
Home duties	36
Student	24
Retired/pension	43
Indigenous
Yes	50*
No	36
Weapon
Yes	61*
No	30
Type of incident
Threat	30*
Attack	48
Physical injury
Yes	54*
No	33
Number of offenders
One	35*
Two	33
Three or more	48
-- fewer than five cases * Series is statistically significant, X2, p<.05 Source: Australian Institute of Criminology, International crime victimisation survey, 2004 [computer file]

Table 6: Rates of reporting burglary by household characteristics (per cent)

	Reported to police
Household income
<$400 per week	78*
$400 or more	85
Time at postcode
<1 year	75*
1 year or more	85
Property loss
Yes	88*
No	67
Value of property loss
<$1,000	75*
$1,000 or more	95
* Series is statistically significant, X2, p<.05 Source: Australian Institute of Criminology, International crime victimisation survey, 2004 [computer file]

Policy implications

Since the first victimisation surveys were undertaken in the 1970s, and as they have continued to be developed and methodologies enhanced, they have enriched our understanding of the nature and extent of crime in our communities and citizens' reaction to crime. With ongoing collections it is possible to see changes over time. A key finding from the Australian component of the 2004 ICVS survey is that there appears to have been a decline in actual crime and levels of fear since the last survey was conducted four years ago.

This survey has provided important information that can inform cross-sector, government and community efforts to prevent and reduce fear and criminal victimisation. For example:

• Risk of victimisation: risk of personal crime is associated with being unmarried, unemployed, and having relatively high residential mobility, and an active lifestyle outside the home in the evenings. In addition, Indigenous status and speaking only English at home raise the risk of assault or threat. This supports the need for a wide range of crime prevention measures and approaches related to reducing opportunities for crime and the vulnerability of specific groups in the community.
• Repeat victimisation: one of the most important risk factors of victimisation is prior victimisation. With this knowledge, police have a critical role in preventing the recurrence of crimes by helping citizens identify and eliminate vulnerabilities that can lead to a repeat experience of the same or another type of crime.
• Unreported crime: a substantial number of unreported crimes involve loss or harm to victims and victims make the decision not to involve the police for personal reasons. However, a small but significant proportion did not report assaults for reasons such as fear of retaliation by the offender (see Johnson 2005). An awareness of this can help police and support services reach out to reluctant victims and prevent a repeat of the crime.
• Factors related to fear: these include personal vulnerability that accompanies being female, young, belonging to a minority group who speaks a language other than English at home, and experiences of crime victimisation. Identifying different levels of fear in the community and the reasons for their fear can help local communities develop strategies to address fear of crime, including alterations to the built environment or reducing social or physical disorder.

Acknowledgement

The ICVS was funded by the Australian Government Attorney-General's Department and the Australian Government Department of Immigration and Multicultural and Indigenous Affairs.

References

• Australian Bureau of Statistics 2003. Crime and safety Australia cat no 4509.0. Canberra: Australian Bureau of Statistics
• Australian Institute of Criminology 2004. Australian crime: facts and figures 2004. Canberra: Australian Institute of Criminology
• Carcach C & Makkai T 2003. The Australian component of the 2000 international crime victims survey (ICVS). Technical & background paper no 3. Canberra: Australian Institute of Criminology
• Johnson H 2005. Crime victimisation in Australia: key results of the 2004 international crime victimisation survey. Research and public policy series no 64. Canberra: Australian Institute of Criminology
• Mihorean K, Besserer S, Hendrick D, Brzozowski JA, Trainor C & Ogg S 2001. A profile of criminal victimisation: results of the 1999 General social survey cat no 85-553-XIE. Ottawa: Statistics Canada
• Van Kesteren J, Mayhew P & Nieuwbeerta P 2000. Criminal victimisation in seventeen industrialised countries. The Hague: Ministry of Justice of the Netherlands

Holly Johnson is Visiting Fellow at the Australian Institute of Criminology

Disclaimer: This research paper does not necessarily reflect the policy position of the Australian Government