Francis Markham, Australian National University; Bruce Doran, Australian National University, and Martin Young
Our recently published research suggests a link between the number of poker machines in an area and levels of domestic violence.
We compared publicly available data on poker machine numbers with police-recorded domestic violence incidents in Victoria between 2005 and 2014. Specifically, we compared the number of pokies and pokie venues in each Victorian postcode to the number of “family violence incidents” recorded by the police, and the number of those incidents that led to a formal charge of assault.
Police-recorded domestic violence is associated with many other social indicators apart from poker machines. For that reason, we statistically adjusted for other local indicators. These included:
- economic gender inequality;
- social and economic disadvantage;
- the proportion of people from an English-speaking background;
- the average number of children per woman;
- the proportion of residents who identify as Indigenous;
- the median age; and
- geographic remoteness.
What we found
We found a statistically significant correlation between poker machine density and police-recorded domestic violence rates among postcodes. This relationship existed even after accounting for the contextual factors listed above.
In terms of relative risk, the police recorded 20% fewer family violence incidents and 30% fewer domestic violence assaults when postcodes with no poker machines were compared with postcodes with at least 75 pokies per 10,000 people.
Similar patterns were evident when the number of poker machine venues was analysed instead of the number of poker machines.
These correlations are stronger than we expected. However, it is important to note they account for only a small part of the variation in police-recorded domestic violence rates between postcodes.