Context
My audience for this data story is an advertising agency working with government policy maker to design a targeted advertising campaign to reduce the road toll.
Since record-keeping in the Australian Road Deaths Database began in 1989, 48,896 people have died on our roads or in related incidents. A staggering number.
With the number of deaths beginning to plateau in recent years, authorities are looking at ways keep the trend declining through ongoing education campaigns such as Toward Zero.
Perspective
Safety vs Behaviours
Both the automotive and airline industries have adopted and improved technology to improve safety. For cars, improvements in road quality and engineering have improved overall safety.
User behaviours however, separates these two groups: Commercial airline pilots undergo thousands of hours of training (and are older by this time), and are regularly tested for drug and alcohol policy compliance.
Drivers by comparison are younger, undergo considerably less training an are less subject to drug and alcohol controls. Until reckless behaviours of drivers are addressed, the road toll will never reach zero.
Identifying behaviours
The following visualisations explore fatalities after the year 2000, by time of day and day of week to illustrate behaviours that may be useful in developing an effective advertising campaign.
Men & Women
The first split is by gender. Male deaths make up 71.3% of all deaths recorded.
Young & Old
The next split is by age, where I’ve separated the male fatalities into under 25 and over 25.
The accident
Let’s try to understand this further by looking at the nature of the incident. We can begin to form a view of responsibility by splitting the fatalities of young men between single or multiple vehicle incidents. This view allows us to rule out the influence of other vehicles in the accident.
Roles
Unlike previous indicators where there was a distinction between the two groups, here the fatalities are shared between drivers and passengers. The campaign could include an element of responsibility for both driver and passenger lives.
Speed
I’ve split the remaining dataset into under and over 60km/h (the usual residential area speed limit). Interestingly, the bulk of fatalities have occurred in areas where the speed limit was below 60. To end up in a fatal accident in a 60 zone suggests speed was a contributing factor.
This analysis has revealed that young men speeding after a night out in residential areas, are putting themselves at high-risk of ended up in a fatal accident. An effective campaign would also highlight the responsibility of drivers not just for their own lives, but for their passengers too. Bring on the driverless cars!