Personal Safety, Australia methodology

Latest release
Reference period
2021-22 financial year
Released
15/03/2023
Next release Unknown

Overview

This publication contains results from the 2021-22 Personal Safety Survey (PSS), conducted by the Australian Bureau of Statistics (ABS) throughout Australia from March 2021 to May 2022.

The survey collected information from men and women aged 18 years and over about the nature and extent of violence experienced since the age of 15. It also collected detailed information about men's and women's experiences of:

  • Violence, emotional abuse, and economic abuse by a cohabiting partner
  • Sexual harassment in the last 12 months
  • Stalking
  • Abuse and witnessing parental violence during childhood
  • Feelings of general safety

The survey also collected a standard set of information about respondents, including age, sex, country of birth, main language, employment, education, disability status and income.

The PSS was previously conducted by the ABS in 2016, 2012 and 2005. The survey is adapted from the design of the Women's Safety Survey (WSS) which was conducted in 1996. This release includes some data comparisons with results from previous surveys where appropriate.  

For detailed information about how the PSS was conducted refer to the Personal safety survey: User guide 2021-22.

Acknowledgments

The ABS acknowledges the lives and experiences of people affected by violence and abuse who are represented in this report, and would like to thank respondents for their participation in the survey. Their participation has contributed valuable information that will help to inform public debate about violence and shape further development of policies and programs aimed at reducing the prevalence of violence.

The ABS would also like to acknowledge the support and input of the Department of Social Services (DSS), which provided funding for the 2021-22 Personal Safety Survey under the National Plan to Reduce Violence Against Women and their Children (updated now to the National Plan to End Violence against Women and Children 2022-32), as well as the Survey Advisory Group, who provided the ABS with advice on the priority information to be collected and on some aspects of survey methodology. Members of this group included representatives from state and territory and Commonwealth Government departments, crime research agencies, service providers and relevant academics.

Data collection

Survey content

The survey collected the following content:

  • Demographic information (current at the time of interview), including age, sex, sexual orientation, country of birth, ancestry, main language spoken, and marital status
  • Socio-economic information (current at the time of interview), including labour force status, current study and educational attainment, financial stress, and personal and partner income
  • Household information (current at the time of interview), including household composition, tenure type, landlord type, number of bedrooms, and household income
  • Language, education, and employment of current partner
  • General health and wellbeing, including life satisfaction, self-assessed health status, disability, social connectedness, and general safety
  • Migrant and visa status
  • Defence Force service
  • Violence since the age of 15
  • Cohabiting partner violence
  • Cohabiting partner emotional and economic abuse
  • Sexual harassment in the last 12 months
  • Stalking since the age of 15
  • Sexual and physical abuse before the age of 15
  • Witnessing parental violence before the age of 15

Scope

The scope of the 2021-22 Personal Safety Survey was persons aged 18 years and over residing in private dwellings across Australia (excluding very remote areas). Interviews were conducted with one randomly selected person aged 18 years or over who was a usual resident of the selected household.

Private dwellings are:

  • Houses
  • Flats
  • Home units
  • Any other structures used as private places of residence at the time of the survey

Usual residents are people who usually live in a particular dwelling and regard it as their own or main home. People usually residing in non-private dwellings, such as hotels, motels, hostels, hospitals, nursing homes, or short-stay caravan parks were not in scope.

Both urban and rural areas in all states and territories were included in the survey, except for very remote areas of Australia. The following groups were also excluded from the scope of the survey:

  • Visitors at a dwelling whose usual place of residence is Australia (as they would have their chance of selection at their usual residence)
  • Overseas visitors intending to stay in Australia for less than 12 months
  • Non-Australian diplomats, non-Australian diplomatic staff, and non-Australian members of their household
  • Members of non-Australian defence forces stationed in Australia and their dependants
  • People who usually reside in non-private dwellings
  • Households where all residents are aged less than 18 years

Sample design

The sample size, distribution, and method of selection for the 2021-22 PSS were based on several factors, including:

  • Key estimates required to be produced from the survey
  • Level of disaggregation and accuracy at which these key survey estimates were required
  • Costs and operational constraints of conducting the survey

The aim of the survey was to produce key estimates of interest with an acceptable level of quality. This included sexual, physical and partner violence and partner emotional abuse estimates. Each of these key estimates were then required to be disaggregated for:

  • Women: at the national and state/territory level
  • Men: at the national level

Impacts of the COVID-19 pandemic on the survey

The PSS was originally scheduled for enumeration in 2020. However, the COVID-19 pandemic and associated government responses resulted in multiple postponements to data collection activities.

These disruptions were experienced across the ABS statistical program and led to a number of adjustments to the household survey schedule and collection approach. Resulting adjustments included postponing surveys, reducing content, introducing digital collection channels, and reducing sample sizes to balance priorities across the survey program.

For the PSS, the decision was taken to reduce and redesign the approached sample to maximise the ability to report against key information requirements. While the reduced sample maintains the quality of headline indicators, it has had some impact on the range of data able to be produced.

The PSS response rate was also impacted by important design features which the ABS implemented to help ensure the safety of respondents and interviewers. These included:

  • The partially voluntary nature of the survey;
  • The requirement for interviews to be conducted in a private interview setting;
  • The decision not to conduct proxy interviews for the voluntary/sensitive component of the survey.

A final response rate of 52.2% was achieved, with 11,905 persons completing the entire questionnaire (both compulsory and voluntary components) nationally – 9,832 women and 2,073 men. The final data was benchmarked and weighted to be representative of the in-scope population.  

Collection method

Enumeration ran between March 2021 and May 2022. The data was collected via personal face-to-face interviews (computer-assisted personal interview or CAPI), including the option of self-interview (computer-assisted self-interview or CASI) where the respondent completes the interview themselves using the interviewer’s laptop, instead of having the interviewer ask the questions aloud.

Telephone interviewing (computer-assisted telephone interview or CATI) was developed as a contingency during initial nationwide lockdowns. Small-scale testing was carried out in two stages in 2021 before being approved for use in the field in early 2022. Supporting telephone enumeration was critical to achieving collection outcomes, while ensuring interviewers and respondents remained safe and abided by public health guidelines. Approximately one-third of the sample were interviewed by telephone.

While a new mode of data collection (CATI) was introduced in the 2021-22 survey for the first time, a review of prevalence data by sex and mode did not identify any systematic mode impacts, and the results produced were broadly consistent across the three different data collection modes used to enumerate the survey. Any differences that were found between the modes were small, and consistent with the expectation that there will be some natural variability in results across the mode types. All care was taken to preserve comparability with previous PSS results to enable time series comparisons.

The first component of the interview involved collecting demographic and household characteristics information. Once questions regarding a person’s experience of violence were reached, respondents were informed of the nature of the upcoming questions and their permission to continue with the interview was sought (referred to as the opt-out point). Respondents had the option to end the interview at any stage, even if they chose to continue beyond the opt-out point. 

Information was collected by specially trained ABS interviewers, who were provided with detailed instruction manuals about the survey content and the procedures to be followed. For the 2021-22 PSS, male interviewers were used to interview male respondents only.

Training of the ABS interviewers was conducted in a variety of formats:

  • Prior to the COVID-19 pandemic restrictions, interviewers attended a two day face-to-face comprehensive training workshop, which also included sensitivity training delivered by a psychologist.
  • During COVID-19 pandemic restrictions, interviewers were trained via online video conference, video presentations, and one-on-one virtual practice sessions. They were also extensively trained in conducting Computer-Assisted Telephone Interviews (CATI) and were given instruction manuals to assist. This training highlighted the importance of only conducting a CATI when the selected person was in a private setting, as was the requirement when conducting face-to-face Computer-Assisted Personal Interviews (CAPI).
  • After COVID-19 pandemic restrictions had eased, interviewers in states/territories that experienced long-term lockdown restrictions were given a one-day online refresher training course.

Data processing

Estimation methods

Estimation is a technique used to produce information about a population of interest, based on a sample of units (i.e. persons) from the population. Each sample unit in the 2021-22 PSS has a person weight. Information for sampled persons is multiplied by the weights to produce estimates for the total in-scope population.

Weighting

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).

Benchmarks

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 any over- or under-enumeration of particular categories of persons/households, which may occur due to either the random nature of sampling or non-response.

The 2021-22 PSS survey estimates were benchmarked to the estimated resident population of Australia (excluding very remote areas) aged 18 years and over who were living in private dwellings as at September 2021.

Reliability of the estimates

The estimates in this publication are based on information obtained from a sample survey. Any data collection may encounter factors, known as non-sampling error, which 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 the results 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

Non-sampling error is any factor (other than those related to sample selection) that causes the data to deviate from the true population value and can occur at any stage throughout the survey process. Examples include:

  • selected people that do not respond (e.g. refusals, non-contact) 
  • questions being misunderstood
  • responses being incorrectly recorded
  • errors in coding or processing the survey data.

Sampling error

Sampling error is the expected difference that can occur between the survey estimates and the values that would have been produced if the whole population had been surveyed. Sampling error is the result of random variation and can be estimated using measures of variance in the data.

Standard error

One measure of sampling error is the standard error (SE). There are about two chances in three that an estimate will differ by less than one SE from the value that would have been obtained if the whole population had been included. There are about 19 chances in 20 that an estimate will differ by less than two SEs.

Relative standard error

The relative standard error (RSE) is obtained by expressing the standard error as a percentage of the estimate.

\(RSE\%=(\frac{SE}{estimate})\times100\)

Only estimates with RSEs less than 25% are considered reliable for most purposes. Estimates with higher RSEs (between 25% and 50%) have been included in the publication but are flagged to indicate that they should be used with caution. Estimates with RSEs higher than 50% are considered unreliable for most purposes and have also been flagged. RSEs for these estimates are not published.

Calculating measures of error

Proportions and percentages 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 the numerator (x) is a subset of the denominator (y):

\({RSE}\left(\frac{x}{y}\right) \approx \sqrt{[R S E(x)]^{2}-[R S E(y)]^{2}}\)

When calculating measures of error, it may be useful to convert RSE to SE. This allows the use of standard formulas involving the SE. The SE can be obtained from RSE using the following formula:

\(S E(y)=\frac{R S E(y) \times y}{100}\)

Calculating differences

The difference between two survey estimates (counts or percentages) can also be calculated from published estimates. Such an estimate is also subject to sampling error. The sampling error of the difference between two estimates depends on their SEs and the relationship (correlation) between them. An approximate SE of the difference between two estimates (x-y) may be calculated by the following formula:

\(S E(x-y) \approx \sqrt{[S E(x)]^{2}+[S E(y)]^{2}}\)

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.

Significance testing

When comparing estimates between survey collections over time or between populations (e.g. men and women) within a survey, it is useful to determine whether any apparent differences are 'real' differences or simply the product of differences between the survey samples (sampling variability). 

One way to examine this is to determine whether the difference between the estimates is statistically significant. This is done by calculating the standard error of the difference between two estimates (x and y) and using that to calculate the test statistic using the formula below. 

\(\left(\frac{|x-y|}{S E(x-y)}\right)\)

where:

\(S E(y)=\frac{R S E(y) \times y}{100}\)

If the value of this test statistic is greater than 1.96, there is good evidence of a statistically significant difference (at 95% confidence levels) between the two populations with respect to the characteristic being compared. Otherwise, it cannot be stated with confidence that there is a real difference between the populations.

Significant differences identified by the ABS have been annotated with a footnote in selected published 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 time points, or undertake significance testing using the formula provided to determine whether there is a statistically significant difference between any two estimates.

Data output

Release strategy

This first release of data focuses on high-level violence prevalence rates at the national level for men and women (lifetime and 12-month rates) and at the jurisdictional level for women only (lifetime, 12-month, and 24-month rates). Subsequent publications and articles containing more detailed and topic-focused data will continue to be released throughout 2023.

Data cubes (spreadsheets released as part of the suite of publication products) present tables of estimates, proportions, and their associated relative standard error, and are available for download in the Data downloads section of the main publication page.

Confidentiality

The Census and Statistics Act 1905 authorises the ABS to collect statistical information and requires that information is not published in a way that could identify a particular person or organisation. The ABS has an obligation to ensure that information about individual respondents cannot be derived from published data.

To minimise the risk of identifying individuals in aggregate statistics, a technique is used to randomly adjust cell values. This technique is called perturbation. 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.

Comparing the data

Comparability with previous Personal Safety Surveys

The topics and information collected in the 2021-22 Personal Safety Survey remained largely consistent with previous iterations of the survey. As a similar methodology and violence definitions have been used across all Personal Safety Surveys from 2005 onwards, prevalence rates for experiences in the last 12 months (and up to the last 5 years) are comparable across the survey periods, unless noted otherwise. This has enabled some time series comparisons between 2021-22 data and results from previous surveys to be made in this publication.

Differences in prevalence rates between 2016 and 2021-22 have been tested for statistical significance, and any statistically significant differences are annotated in the data tables and described in the commentary. For more information about statistical significance, refer to the Data Processing section above.

The 2016 survey included some minor changes to definitions to assist with respondents’ understanding of survey questions, and expanded the list of behaviours asked about for some topics (e.g. technology-facilitated behaviours were added to the sexual harassment, stalking, and emotional abuse topics).

The changes to the 2021-22 PSS sample design and mode of data collection in response to the COVID-19 pandemic included:

  • Sample design – Reduced sample size for both men and women
  • Collection mode – The introduction of Computer-Assisted Telephone Interviews (CATI), which gave respondents the option to complete the survey over the phone.

The 2021-22 PSS sample reduction has impacted the availability of some data for men and all persons. The following data have been omitted from the publication tables due to high error values:

  • 12-month prevalence rates by relationship to perpetrator and type of violence for men (with the exception of physical assault)
  • 12-month prevalence rates for all persons (due to the high error values on male estimates)
  • All prevalence data at the jurisdictional level for men and all persons

Comparability with other data sources

The ABS collects and publishes data relating to crime and safety from a variety of sources, including data collected through the annual Crime Victimisation Survey and published in Crime Victimisation, Australia, and administrative data collected from police agencies and published annually in Recorded Crime – Victims. Comparisons of PSS data with data from other sources cannot be readily made due to differences in data collection methods and the concepts and definitions used to measure violence.

Glossary

Show all

Post release changes

05/03/2024

Link to the Personal safety survey: User guide, 2021-22 added to the Overview section. 

Back to top of the page