Media representations of violence against women and their children: Final report

Monday, 6th June 2016

This project aimed to establish the extent and nature of reporting of violence against women by the Australian media to inform future strategies for change.

Using both quantitative (content analysis) and qualitative (critical discourse analysis) methods, the study provided a glimpse into the complexity of reporting practices.

It found that:

  • There is a clear link between media reporting and attitudes and beliefs in relation to violence against women, with audiences’ emotional responses and attributions of responsibility affected by how the media frames news.
  • The vast majority of reporting on violence against women was “incident based”, looking at tragic individual instances, but not exploring the issue in a more depth.
  • The lack of social context in reporting, and thereby the broader public’s understanding of the issue, could be improved by the inclusion of more expert sources, including domestic violence advocates and those with lived experience of violence. Yet half of all sources were drawn from police and the criminal justice system; only 9.9 % of sources were domestic violence advocates /spokespeople; only 8.7 % were survivors.
  • The narrow use of sources contributed to a “murder centric” frame of most reporting, which is no doubt newsworthy but doesn’t necessarily reflect women’s different experiences of violence. 61.8% of incident based reporting was in relation to a homicide. Nearly 75.8 % of reporting focused on physical intimate partner violence, 22.5 % on sexual assault. Other types of violence, including emotional, threats or sexual harassment were all but invisible.
  • Myths and misrepresentations still find their way into reporting. Around 15 % of incident based reporting includes victim blaming, like she was drinking, flirting/went home with the perpetrator, was out alone, they were arguing; 14.8 % of incident based reporting offers excuses for the perpetrator, like he was drinking, using drugs, jealous/seeking revenge, “snapped” or “lost control”.
  • Interestingly, and for the first time, this research  picked up on a tendency to render the perpetrator invisible, with 59.8% of incident based reporting including no information whatsoever about the perpetrator.
  • Choice of language can sometimes be insensitive, for example 17.2 % of newspaper and online headlines were deemed sensationalistic, while 13.3 % of incident based news items used language in the report that was sensationalistic, including excessively gory/or overly sexually explicit detail.
  • Though we know the news media can be a powerful source of information for women looking to leave a violent relationship, only 4.3 % of news reports included help seeking information (1800RESPECT or others).

Download Media representations (full copy high res).pdf (21.19 MB)

Download Media representations (compressed for smaller file size - low print quality).pdf (8.98 MB)

ANROWS Horizons are in-depth reports on empirical research produced under ANROWS’s Research Program.