Family, domestic and sexual violence data in Australia

Notes

1. Emotional abuse occurs when a person is subjected to certain behaviours or actions that are aimed at preventing or controlling their behaviour, causing them emotional harm or fear. These behaviours are characterised in nature by their intent to manipulate, control, isolate or intimidate the person they are aimed at. They are generally repeated behaviours and include psychological, social, economic and verbal abuse. Behaviours included under Emotional abuse has expanded over time to allow for addition of some new content or concepts (for example additional technologically focused behaviours were added), resulting in the list of included behaviours varying between surveys. For this reason, care should be taken when interpreting the results over time. An increase may be reflective of a real-world change or a change in the survey questionnaire.
2. Proportions are based on all adults and are not limited to those who have/had a current or previous partner. 

3. Separate current partner and previous partner counts are not able to be added together to produce a current and/or previous partner total. Where a person has experienced both emotional abuse by a current partner and by a previous violence, they are counted separately for each type of violence they experienced but are counted only once in the aggregated total.

4. The PSS defines partner as a person the respondent lives with, or lived with at some point in a married or de facto relationship. A current partner is a person who, at the time of the survey, was living with the respondent in a marriage or de-facto relationship, and a previous partner as a person who lived with the respondent at some point in a marriage or de facto relationship, but who was no longer living with the respondent at the time of the survey.
5. The PSS collects information from women and men aged 18 years and over.
6. Survey data, obtained from a sample of the population, is subject to sampling error. Where estimates are subject to a level of sampling error too high for general use, they are not included in visualisations, but are included in data tables, with caveats.
7. The observed value of a rate may vary due to chance even where there is no variation in the underlying value of the rate. The margin of error is the largest possible difference (due to sampling error) that could exist between the estimate and what would have been produced had all persons been included in the survey. Confidence intervals—constructed by taking the estimate plus or minus the MoE— can be used to provide an approximate indication of the true differences between rates. If the confidence intervals do not overlap, the difference can be said to be statistically significant. However, statistically significant differences are not necessarily the same as differences considered to be of practical importance. Small differences that have practical importance may be found to be not statistically significant as they are below the threshold the significance test can reliably detect.

8. Changes over time may indicate a change in the underlying rate of violence, a change in the propensity to report this violence, or both.

9. For more information see Methods, Glossary and Data sources.

Sources

ABS 2016 Personal Safety Survey  

Next expected: 2022