The Good news, the bad news and then the statistics

By Jane Gilmore

Violence in Australia has been steadily decreasing for the last 20 years. The homicide rate, for example, has dropped by more than 60 percent since 1993. Kidnapping and abductions have halved over the same time. National data on assault is more difficult to find, but state-based data indicates assaults reported to police have also decreased.

Declining rates of violence rarely make news. Stories about violence or misused statistics (the sort of thing that will scare your grandmother) get lots of clicks and “what wrong with the world these days” comments on social media. A story that tells people they are safer than they used to be isn’t so appealing. It doesn’t hit the viral target of fear, racism and nostalgia like a “You’re All In Danger From African Gangs” headline will do.

Often the main problem with crime and violence statistics is that journalists may not understand what the data is saying. Context is everything.

Take the homicide rate decrease I quoted above. This comes from the Australian Bureau of Statistics (ABS) data on crime victimisation. The homicide rate in 1993 was 4 victims for every 100,000 people. By 2019 it was 1.6. This is clearly a significant and welcome drop, but I have some questions.

What percentage of the numbers in 1993 included drunk drivers who killed someone? Has reduced rates of drink driving been a significant factor in reducing homicides? How much of the decrease is due to improved police and ambulance response and better medical care in emergency rooms? Has the number of missing persons increased over that time? What happened to the rate of family violence homicides?

There’s no data I can access to answer those questions. Finding the answers would take a great deal more resources than most journalists have available to them for a single story. But it demonstrates the difficulties in relying on statistics without interrogating the data.

The nature of statistics means we always need to look at them in context. What data is collected? How is it collected? What factors could skew the results and how? What other data is available that supports or contradicts the findings? What are the anomalies and what are the explanations for those anomalies? What are the trends over time and are they consistent?

Crime data (such as the “African Gangs!” example) can be misleading. It tells us what crimes are reported to and recorded by police. It doesn’t tell us if a particular blip in the data is in response to a media blitz – for example the 300 percent increase in family violence reporting that happened in Victoria after Christine Nixon and Key Lay made it a key priority for police and the Royal Commission brought it into the public spotlight. It doesn’t tell us about special taskforces (“African Gangs!”) that might lead to increase reports and charges. It doesn’t tell us about changes over time periods that are too small to be significant but still get turned into a headline.

The Personal Safety Survey is another dataset that is frequently misunderstood. It’s where we get the numbers so often quoted in articles about men’s violence against women.

  • 1 in 3 Australian women has experienced physical violence since the age of 15.
  • 1 in 5 Australian women has experienced sexual violence since the age of 15.
  • 1 in 4 Australian women has experienced physical or sexual violence by current or former intimate partner since age 15.

The Personal Safety Survey is conducted by the ABS. It is robustly designed, and the survey is carried out by ABS staff who have completed a two-day training session. There are, however, some issues with it that could skew the results. Of the more than 21,000 households where a person was approached about completing the survey, just under 10,000 refused or didn’t finish the full questionnaire. Some of those were due to illness, death or language problems but nearly 40 percent were people who did not want to continue with the survey after hearing details about the sort of questions they would be asked. There’s no way to know how many of them were people who have had (or are having) experiences of violence that would make such questions intolerable, but it seems likely that at least some of them would opt out for exactly that reason.

The Personal Safety Survey asks people about their perception and understanding of the violence they have experienced. This is key to understanding what the results tell us about abusive relationships.

Coercive control is a relatively new term, but it describes a dynamic of abuse that people who work in the domestic violence sector have understood for a long time. One of the characteristics of coercive controlling relationships is that both the victim and the preparator often do not perceive it as violent. Victims are groomed, manipulated and gaslit into believing they are the cause of any problem in the relationship.

Jess Hill, author of See What You Made Me Do, says the majority of victims of coercive control do not perceive the relationship as abusive. “They may, in fact, regard themselves as the strong one – the only one who can help a troubled man face his demons,” she says.

Men who use coercive control against their partners often believe themselves to be the victim and are keen to present themselves as the good man being damaged by an unstable and unfaithful woman. Given that dynamic, it’s not difficult to understand how women in an abusive relationship might play down the violence they experience, and abusive men might exaggerate.

The ABS is not at fault for reporting this data. Their job is to ask the questions and record the answers. But the complex distortion of perception that is an integral part of coercive control means the results may be skewed in ways the survey cannot establish. The rate at which traumatised women opt out and abusive men opt in is also beyond their ability to control or measure.

None of this is to say that the data from the Personal Safety Survey is useless. It isn’t. it gives us a very good indication of how people perceive the violence they experience but it needs to be understood as exactly that – an indication, not a set of verified facts.

It’s likely, but not provable, that abusive relationships are actually far more common than the Personal Safety Survey indicates.

The most reliable data I can find does indicate violence in Australia is decreasing. That’s the good news. The bad news is that abusive relationships don’t seem to have changed very much. The statistics, as always, are open to interpretation.

Jane Gilmore was the founding editor of The King’s Tribune. She is now a freelance journalist and author, with a particular interest in feminism, media and data journalism and has written for The Guardian, The Sydney Morning Herald, The Age, The Daily Telegraph, The Saturday Paper and Meanjin, among many others. Jane has a Master of Journalism from the University of Melbourne, and her book FixedIt: Violence and the Representation of Women in the Media was published by Penguin Random House in 2019.

Support quality journalism.

As an inkl member you can directly support the work of journalists like Jane Gilmore, while also getting access to 100+ publications like Foreign Affairs, The Independent, The Economist, Financial Times and Bloomberg.

As part of our commitment to building a sustainable future for journalism, a portion of your monthly inkl membership fee will go directly to Jane for as long as you remain a subscriber.

BECOME A MEMBER

What is inkl?

Important stories

See news based on value, not advertising potential. Get the latest news from around the world.

Trusted newsrooms

We bring you reliable news from the world’s most experienced journalists in the most trusted newsrooms.

Ad-free reading

Read without interruptions, distractions or intrusions of privacy.