Crime Victimisation, Australia methodology

This publication presents results from the Crime Victimisation Survey, a topic on the Multipurpose Household Survey (MPHS) conducted throughout Australia from July 2018 to June 2019. The MPHS, undertaken each financial year by the Australian Bureau of Statistics (ABS), is a supplement to the monthly Labour Force Survey (LFS) and is designed to collect statistics for a number of small, self-contained topics.

This publication covers the Crime Victimisation topic and presents details about the prevalence of a selected range of personal and household crimes, including the socio-demographic characteristics of persons experiencing the selected crimes, experiences of repeat victimisation, and the characteristics of the most recent incident of each crime type experienced. Some estimates from previous iterations of the Crime Victimisation Survey are also included in this publication.

The Crime Victimisation Survey is being conducted again as part of the MPHS for the reference period 2019-20, with results expected to be released in early 2021.

The scope of the Crime Victimisation Survey was restricted to people aged 15 years and over who were usual residents of private dwellings and excludes:

• members of the Australian permanent defence forces;
• certain diplomatic personnel of overseas governments, customarily excluded from Census and estimated resident population counts;
• overseas residents in Australia; and
• members of non-Australian defence forces (and their dependants).
   

Additionally, the 2018-19 MPHS scope excluded:

• persons living in non-private dwellings such as hotels, university residences, boarding schools, hospitals, nursing homes, homes for people living with disabilities, and prisons; and
• persons resident in the Indigenous Community Strata (ICS).
   

The scope for MPHS included households residing in urban, rural, remote and very remote parts of Australia, except the ICS. In the LFS, rules are applied which aim to ensure that each person in coverage is associated with only one dwelling, and hence has only one chance of selection in the survey. See Labour Force, Australia (cat. no. 6202.0) for more detail.

The Crime Victimisation Survey is one of a number of small, self-contained topics on the Multipurpose Household Survey (MPHS), conducted throughout Australia from July 2018 to June 2019. The MPHS is a supplement to the monthly LFS. In 2018–19, the MPHS topics were:

• Crime Victimisation;
• Patient Experiences in Australia;
• Barriers and Incentives to Labour Force Participation;
• Retirement and Retirement Intentions;
• Qualifications and Work; and
• Income (Personal, Partner's, Household).
   

For all topics, general demographic information such as age, sex, labour force characteristics, education and income are also available.

ABS interviewers conducted personal interviews during the 2018-19 financial year for the monthly LFS. Each month, one eighth of the dwellings in the LFS sample were rotated out of the survey and selected for the MPHS. After the LFS had been fully completed for each person in scope and coverage, a usual resident aged 15 years or over was selected at random (based on a computer algorithm) and asked the additional MPHS questions.

In the MPHS, if the randomly selected person was aged 15 to 17 years, permission was sought from a parent or guardian before conducting the interview. If permission was not given, the parent or guardian was asked the questions on behalf of the 15 to 17 year old (proxy interview). Questions relating to sexual assault and the involvement of alcohol or substances in the most recent incident of physical assault and face-to-face threatened assault were not asked of proxy respondents. Only persons aged 18 years and over were asked questions on sexual assault and the involvement of alcohol or substances in the most recent incident of physical assault and face-to-face threatened assault.

Data were collected using Computer Assisted Interviewing (CAI), whereby responses were recorded directly onto an electronic questionnaire in a notebook computer, with interviews conducted either face-to-face or over the telephone. The majority of interviews were conducted over the telephone.

After taking into account sample loss, the response rate for the Crime Victimisation Survey was 71.8%. In total, information was collected from 28,719 fully responding persons. This includes 477 proxy interviews for people aged 15 to 17 years, where permission was not given by a parent or guardian for a personal interview.

Weighting is the process of adjusting results from a sample survey to infer results for the total in-scope population. To do this, a 'weight' is allocated to each enumerated person. The weight is a value which indicates the number of persons in the population represented by the sample person.

The first step in calculating weights for each unit is to assign an initial weight, which is the inverse of the probability of being selected in the survey. For example, if the probability of a person being selected in the survey was 1 in 600, then the person would have an initial weight of 600 (that is, they represent 600 people).

The initial weights were calibrated to align with independent estimates of the population of interest, referred to as 'benchmarks'. Weights calibrated against population benchmarks ensure that the survey estimates conform to the independently estimated distribution of the population rather than the distribution within the sample itself. Calibration to population benchmarks helps to compensate for over-or-under-enumeration of particular categories of persons/households which may occur due to either the random nature of sampling or non-response.

For household estimates, the MPHS was benchmarked to independently calculated estimates of the total number of households in Australia. The MPHS estimates do not (and are not intended to) match estimates for the total Australian person/household populations obtained from other sources.

The survey was benchmarked to the Estimated Resident Population (ERP) living in private dwellings in each state and territory at December 2018, based on the 2016 Census. People living in Indigenous communities were excluded.

While the LFS benchmarks are revised every five years to take into account the outcome of the five-yearly rebasing of the ERP following the latest Census, the supplementary surveys and MPHS (from which the statistics in this publication are taken) are not. Small differences will therefore exist between the civilian population aged 15 years and over reflected in the LFS and other labour household survey estimates, as well as over time.

Survey estimates of counts of persons are obtained by summing the weights of persons with the characteristic of interest.

To minimise the risk of identifying individuals in aggregate statistics, a technique called perturbation is used to randomly adjust cell values. Perturbation involves a small random adjustment of the statistics and is considered the most satisfactory technique for avoiding the release of identifiable statistics while maximising the range of information that can be released. These adjustments have a negligible impact on the underlying pattern of the statistics.

After perturbation, a given published cell value will be consistent across all tables. However, adding up cell values to derive a total will not necessarily give the same result as published totals.

Perturbation has been applied to Crime Victimisation Survey datasets since 2013–14. Data from previous cycles (2008-09 to 2012–13) have not been perturbed, but underwent a different confidentialisation method to protect the confidentiality of respondents.

All sample surveys are subject to error which can be broadly categorised as either sampling error or non-sampling error.

Sampling error is the difference between the published estimates, derived from a sample of persons, and the value that would have been produced if the total population (as defined by the scope of the survey) had been included in the survey.

Non-sampling error may occur in any statistical collection, whether it is based on a sample or a full count such as a census. Sources of non-sampling error include non-response, errors in reporting by respondents or recording of answers by interviewers, and errors in coding and processing data. Every effort is made to reduce non-sampling error by careful design and testing of questionnaires, training and supervision of interviewers, and extensive editing and quality control procedures at all stages of data processing.

Only data with a relative standard error (RSE) of less than 25% are included in the publication commentary (unless otherwise noted), and any differences between populations and changes over time that are referred to are statistically significant. All data contained in the commentary are available for download as data cubes from the Data downloads section. For more information about relative standard error and statistical significance refer to the Technical Note.

Crime victimisation surveys are best suited to measuring crimes against specific individuals or households. Respondents need to be aware of and recall what happened to them and how it happened, as well as be willing to relate what they know to interviewers.

Not all types of crime are suitable for measurement by household surveys. No reliable information can be obtained about crimes without specific victims, such as trafficking in narcotics. Crimes of which a person may not be aware cannot be measured effectively through a household survey, for example crimes involving deception. It may also be difficult to obtain information about some crimes, such as sexual offences and assault committed by other household or family members, due to the sensitivity of the crime and an increased reluctance to disclose. Some of these crimes may not be fully represented in the data collected. Household survey data exclude crimes against commercial establishments or government agencies.

This survey covered only selected types of personal and household crimes and does not represent all crime in Australia. Personal crimes covered in the survey were physical assault, threatened assault (face-to-face and non-face-to-face), robbery and sexual assault. Household crimes covered were break-in, attempted break-in, motor vehicle theft, theft from a motor vehicle, malicious property damage and other theft.

Information collected in this survey is essentially 'as reported' by respondents and hence may differ from that which might be obtained from other surveys or administrative data sources. This factor should be considered when interpreting the estimates and when making comparisons with other data sources.

There is limited information available in this publication about family and domestic violence. The Crime Victimisation Survey collects information about experiences of personal violence and the relationship between the victim and perpetrator, however this information alone is not sufficient to reliably measure the number of people who have experienced family and domestic violence.

The Crime Victimisation Survey collects incident characteristics information, including relationship to the offender, only for the most recent incident of each type of personal crime experienced in the 12 months prior to interview. This means that not all experiences of personal violence by each relationship type - including current and previous partners - are captured in the survey. In addition, as interviews are conducted by telephone in the respondent’s home, there is no requirement for a private interview setting for the Crime Victimisation Survey (as is the case for the ABS’s Personal Safety Survey). This non-private setting means respondents may be less likely to disclose any experiences of violence by their partner if their partner is present in the home at the time of interview. As a result, the statistics on relationship type available in this publication cannot be used to draw conclusions about the prevalence of family and domestic violence in Australia.

Due to the ongoing relationship between victim and perpetrator, family and domestic violence is often a recurring event, and the protracted nature of this violence cannot be reliably measured within the framework of the Crime Victimisation Survey. Further information about defining and measuring family and domestic violence is available in Defining the Data Challenge for Family Domestic and Sexual Violence (cat. no. 4529.0) and statistics are available in Personal Safety, Australia (cat. no. 4906.0), Directory of Family, Domestic, and Sexual Violence Statistics, 2018 (cat. no. 4533.0), and Recorded Crime - Victims, Australia (cat. no. 4510.0).

The level of victimisation can be measured and expressed in more than one way. The most common measure derived from crime victimisation surveys is prevalence, that is, the number of the relevant population that have experienced a given crime at least once in the reference period. Victimisation rates used in this publication represent the prevalence of selected crimes in Australia, and are expressed as a percentage of the total relevant population. Reporting rates used in this publication are expressed as the percentage of persons/households whose most recent incident of each type of crime had been reported to the police.

The 2018-19 Crime Victimisation Survey is the eleventh in a series of annual Crime Victimisation Surveys conducted by the ABS. The ten previous surveys in this series included the majority of the questions asked in 2018-19. As a similar methodology has been adopted for the surveys, data on the prevalence of personal and household crimes is comparable across the survey periods. This has enabled some time series comparisons to be made in this publication.

The Crime Victimisation Survey series replaced the previous Crime and Safety Surveys and was introduced because of a change to the collection methodology. The new method of collection mainly uses personal telephone interviews of selected respondents. Data collections between 1990 and 2005 required respondents to complete questionnaires by themselves and mail these back to the ABS. This difference in mode of collection and changes to survey content means that Crime Victimisation Survey data collected using the MPHS are generally not directly comparable with data from Crime and Safety Surveys prior to 2008–09.

In 2010-11 significant changes to the 'Area of Usual Residence', 'Capital City' and 'Balance of State/Territory' geographical items were made. From 2008-09 to 2012-13 the Australian Standard Geographical Classification (ASGC) (cat. no. 1216.0) was used to characterise Geographical Classifications. From 2013-14 onwards the Australian Statistical Geography Standard (ASGS) (cat. no. 1270.0.55.001) was used. The ASGS is updated on a five yearly basis in accordance with the Census of Population and Housing. Consequently, the 2018-19 Crime Victimisation Survey saw the introduction of the updated ASGS. More information on this can be found in Australian Statistical Geography Standard (ASGS): Volume 1 - Main Structure and Greater Capital City Statistical Areas, July 2016.

For the 2018-19 reference period, the following data have not been published at the state/territory level:

• Robbery;
• Sexual assault;
• Police reporting data for non face-to-face threatened assault;
• Police reporting data for motor vehicle theft; and
• Contribution of alcohol or another substance data for physical assault and face-to-face threatened assault.
   

A review of data quality found that due to low prevalence and/or high error or volatility, estimates for the above data items are too statistically unreliable for general use. National estimates for these data items are still available in the data cubes, and users are advised to exercise caution when using state and territory level data for these data items from previous iterations of the survey.

Data for selected crimes reported to and recorded by police agencies in a calendar year are available in Recorded Crime - Victims, Australia(cat. no. 4510.0). The Crime Victimisation Survey provides an additional source of data on crime victimisation for the selected crimes, including crime not reported to or detected by police. This survey identifies the nature of this unreported crime, as well as giving information about experiences of repeat victimisation. The information from the survey should be viewed as complementary to police recorded crime statistics.

The terms used for the crimes (such as robbery and physical assault) may not necessarily correspond with the legal or police definitions used. This is because responses obtained in this survey are based on the respondent's perception of the behaviours they experienced. The definitions of terms used in the publication are based on the wording of the questions asked of the respondent and specifications provided to interviewers. Definitions of crime types included in this survey can be found in the Glossary.

The Crime Victimisation Survey collects information on crimes that were reported to police, as well as crimes that went unreported. In this publication, reporting rates are based on whether or not the most recent incident of each crime type experienced in the 12 months prior to interview was reported to police. Interviews were conducted over a 12 month period from 1 July 2018 to 30 June 2019. The actual reference period for a particular respondent was determined by the date of their interview. There is no way of verifying that a crime was reported to police, where the respondent indicated that police were informed.

Another source of variation between the survey results and crimes recorded by police relates to differences in scope. This survey collects information on the personal crimes of physical assault, threatened assault (face-to-face and non-face-to-face), and robbery for all persons aged 15 years and over, and sexual assault for persons aged 18 years and over. In contrast, police statistics include victims of all ages, and any comparisons should take this into consideration. Furthermore, police statistics for a given reference period may include criminal incidents that came to the attention of police during the reference period, but did not occur during it.

Due to differences between collections, caution should be exercised when comparing data from surveys and administrative by-product collections that relate to crime and justice topics. For more information on comparisons between sources, please refer to Measuring Victims of Crime: A Guide to Using Administrative and Survey data, June 2011 (cat. no. 4500.0.55.001).

Caution should be exercised when comparing across ABS surveys and with administrative by-product data that address the access and use of health services. Estimates from the Crime Victimisation Survey may differ from those obtained in other surveys (such as the Personal Safety Survey, National Aboriginal and Torres Strait Islander Health Survey, National Aboriginal and Torres Strait Islander Social Survey, and General Social Survey) due to differences in survey mode, methodology and questionnaire design.

Since the Crime Victimisation Survey is conducted as a supplement to the Labour Force Survey (LFS), data items collected in the LFS are also available in this publication. However, there are some important differences between the two surveys. The LFS had a response rate of over 90% compared to the MPHS response rate of 71.8%. The scope of the Crime Victimisation Survey and the LFS also differ, as outlined in the preceding sections. Due to the differences between the samples, data from the Crime Victimisation Survey and the LFS are weighted separately. Variances may therefore be found in the estimates for those data items collected in the LFS and published as part of the Crime Victimisation Survey.

When interpreting data from the 2018-19 MPHS, consideration should be given to the representativeness of the survey sample in relation to the entire in-scope population. This is affected by the response rate and scope and coverage rules. For example, people living in boarding houses, refuges or on the streets are excluded from this survey and may experience different levels of victimisation than those surveyed who live in private dwellings.

Equivalised weekly household income is household income adjusted by the application of an equivalence scale to facilitate comparison of income levels between households of differing size and composition, reflecting that a larger household would normally need more income than a smaller household to achieve the same standard of living. Using an equivalising factor for household income enables the direct comparison of the relative economic well-being of households of different size and composition (for example, lone person households, families and group households of unrelated individuals).

For more information about equivalised weekly household income see Household Income and Wealth, Australia (cat. no. 6523.0) and Survey of Income and Housing, User Guide, Australia (cat. no. 6553.0).

Socio-Economic Indexes for Areas (SEIFA) is a classification developed by the ABS that ranks areas in Australia according to relative socio-economic advantage and disadvantage. SEIFA uses a broad definition of relative socio-economic advantage and disadvantage in terms of people's access to material and social resources, and their ability to participate in society.

The indexes are based on information from the five-yearly Census, and each index summarises a different aspect of the socio-economic conditions of people living in an area. Every geographic area in Australia is given a SEIFA number which shows how disadvantaged or advantaged that area is compared with other areas in Australia.

The Crime Victimisation Survey uses two indexes from the 2016 Socio-Economic Indexes for Areas (SEIFA) – the Index of Relative Socio-Economic Advantage and Disadvantage; and the Index of Relative Socio-Economic Disadvantage. These measures are derived from Census variables related to income, educational attainment, unemployment, occupational skill level and whether a dwelling has a motor vehicle.

For more detail, see the following:

• SEIFA Basics: Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia (cat. no. 2033.0.55.001)
• SEIFA 2016 Technical paper available from the Data downloads section: Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia (cat. no. 2033.0.55.001)
   

Australian geographic data are classified according to the Australian Statistical Geography Standard (ASGS): Volume 1 - Main Structure and Greater Capital City Statistical Areas, July 2016 (cat. no. 1270.0.55.001). Remoteness areas are classified according to the Australian Statistical Geography Standard (ASGS): Volume 5 - Remoteness Structure, July 2016 (cat. no. 1270.0.55.005).

Country of birth data are classified according to the Standard Australian Classification of Countries (SACC), 2016 (cat. no. 1269.0).

Education data are classified according to the Australian Standard Classification of Education ASCED, 2001 (cat. no 1272.0). The ASCED is a national standard classification which can be applied to all sectors of the Australian education system including schools, vocational education and training and higher education. The ASCED comprises two classifications: Level of Education and Field of Education.

Data cubes containing all tables for this publication in Excel spreadsheet format are available from the Data downloads section. The spreadsheets present tables of estimates and proportions, and their corresponding relative standard errors (RSEs).

As well as the statistics included in this and related publications, the ABS may have other relevant data available on request. Subject to confidentiality and sampling variability constraints, tables can be tailored to individual requirements. A list of data items from this survey is available from the Data downloads section. All enquiries should be directed to the National Information and Referral Service on 1300 135 070, or email client.services@abs.gov.au.

ABS surveys draw extensively on information provided by individuals, businesses, governments and other organisations. Their continued cooperation is very much appreciated and without it, the wide range of statistics published by the ABS would not be available. Information received by the ABS is treated in strict confidence as required by the Census and Statistics Act 1905.

The ABS Privacy Policy outlines how the ABS will handle any personal information that you provide to the ABS.

The estimates in this publication are based on information obtained from a sample survey. Errors in data collection or processing, known as non-sampling error, can impact on the reliability of the resulting statistics. In addition, the reliability of estimates based on sample surveys are also subject to sampling variability. That is, the estimates may differ from those that would have been produced had all persons in the population been included in the survey. This is known as sampling error.

Non-sampling error may occur in any statistical collection, whether it is based on a sample or a full count such as a census. Sources of non-sampling error include non-response, errors in reporting by respondents or recording of answers by interviewers, and errors in coding and processing data.

It is not possible to quantify non-sampling error, however, every effort is made to reduce non-sampling error by careful design and testing of questionnaires, training and supervision of interviewers, and extensive editing and rigorous quality control procedures at all stages of data processing.

Sampling error refers to the difference between an estimate obtained from surveying a sample of persons, and the result that would have been obtained if all persons had been surveyed.

One measure of sampling error is given by the standard error (SE), which indicates the extent to which an estimate might have varied by chance because only a sample of persons was surveyed. There are about two chances in three (67%) that a sample estimate will differ by less than one SE from the number that would have been obtained if all persons had been surveyed, and about 19 chances in 20 (95%) that the difference will be less than two SEs.

In this publication, the standard error of the estimate is expressed as a percentage of the estimate, known as the relative standard error (RSE), which is a useful measure as it indicates the size of the error relative to the estimate.
 

Only estimates (counts or percentages) with an RSE of less than 25% are considered sufficiently reliable for most analytical purposes. However, estimates with an RSE over 25% are also published. Estimates with an RSE in the range 25% to 50% are less reliable and should be used with caution, while estimates with an RSE greater than 50% are considered too unreliable for general use.

The Excel files available from the Data downloads section contain all the data tables produced for this release, including all estimates and their corresponding RSEs. All cells in the Excel spreadsheets containing an estimate with an RSE of 25% or greater are annotated with asterisks, indicating whether the RSE of the estimate is in the range 25% to 50% (single asterisk *) or is greater than 50% (double asterisk **).

For more details see What is a Standard Error and Relative Standard Error, Reliability of estimates for Labour Force data.

Standard error (SE) can be calculated using the estimate (count or percentage) and the corresponding RSE. For example, if the estimated number of persons who experienced physical assault in the last 12 months was 462,200, with a corresponding RSE of 5.0%, the SE (rounded to the nearest 100) is calculated by:
 

Therefore, there is about a two in three chance that the result that would have been obtained had all persons been included in the survey falls within the range of one standard error below to one standard error above the estimate (439,100 to 485,300), and about a 19 in 20 chance that the result would have fallen within the range of two standard errors below to two standard errors above the estimate (416,000 to 508,400). This example is illustrated in the diagram below:

Proportions formed from the ratio of two estimates are also subject to sampling error. The size of the error depends on the accuracy of both the numerator and the denominator. A formula to approximate the RSE of a proportion is given below. This formula is only valid when x is a subset of y:
 

As an example, if 86,600 persons experienced physical assault by an intimate partner, representing 29.5% of all persons who experienced physical assault by a known person (293,600); and if the RSE for the number of persons experiencing physical assault by an intimate partner is 7.7% and the RSE for the number of persons experiencing physical assault by a known person is 6.7%; then, applying the above formula, the RSE of the proportion is:
 

Using the formula given above, the standard error (SE) for the proportion of persons who experienced physical assault by an intimate partner (as a proportion of those who experienced physical assault by a known person) is 1.1% (0.038 x 29.5). There are about two chances in three that the true proportion of persons who experienced physical assault by an intimate partner (as a proportion of those who experienced physical assault by a known person) is between 28.4% and 30.6%, and 19 chances in 20 that the true proportion is between 27.3% and 31.7%.

The difference between two survey estimates (counts or percentages) is also subject to sampling error, and can therefore be measured using standard error. The standard error of the difference between two estimates is determined by the individual standard errors of the two estimates and the relationship (correlation) between them. An approximate standard error of the difference between two estimates (x,y) can be calculated using the following formula:
 

While this formula will only be exact for differences between separate and uncorrelated characteristics or sub populations, it provides a good approximation for the differences likely to be of interest in this publication.

The difference between two survey estimates can be tested for statistical significance, in order to determine the likelihood of there being a real difference between the populations with respect to the characteristic being measured. The standard error of the difference between two survey estimates (x and y) can be calculated using the formula in the preceding section. This standard error is then used in the following formula to calculate the test statistic:
 

If the value of the test statistic is greater than 1.96, then this supports, with a 95% level of confidence, a real (i.e. statistically significant) difference between the two populations with respect to the characteristic being measured. If the test statistic is not greater than 1.96, it cannot be stated with a 95% level of confidence that there is a real difference between the populations with respect to that characteristic.

The following survey estimates have been significance tested to determine whether any differences are statistically significant:

• Annual changes between 2017–18 and 2018–19 in personal and household crime victimisation rates (Tables 4c and 6c);
• Annual changes between 2017–18 and 2018–19 in personal and household crime reporting rates (Tables 5c and 7c);
• Annual changes between 2017-18 and 2018-19 in the proportion of persons who believed alcohol or any other substance contributed to their most recent incident of physical assault and face-to-face threatened assault (Table 8c);
• Differences between state and territory personal and household crime victimisation rates and equivalent national victimisation rates for 2018–19 (Tables 2, 3, 4c, and 6c); and
• Differences between state and territory personal and household crime reporting rates and equivalent national reporting rates for 2018–19 (Tables 2, 3, 5c, and 7c).
   

Significant differences have been annotated with a footnote in the above tables. In all other tables which do not show the results of significance testing, users should take RSEs into account when comparing estimates for different populations, or undertake significance testing using the formula provided to determine whether there is a statistically significant difference between any two estimates.

Only data with a relative standard error (RSE) of less than 25% are included in the publication commentary, unless otherwise indicated, and any differences between populations and changes over time that are referred to are statistically significant. All data contained in the commentary are available for download as data cubes from the Data downloads section.