Yesterday the ABS released a report called Crime Victimisation 2008-09, which contains the results of a survey on crimes and perceptions of crime that have been experienced by over 25,000 people the ABS surveyed between July 2008 and June 2009.

One of the more unusual parts of the report was a question on perceptions of crime, specifically the perceived crime problems that are manifest in the local neighbourhood. If we use the proportion of people that believe various types of crime are a problem in their neighbourhood, this is what the results come in as, broken down by State (click to expand).


The usual suspects pop up here, starting with those perennial favourites of politicians with nothing else to say – the dreaded hooning (politely called “dangerous/noisy driving” by the good folks at the ABS) and vandalism/graffiti.

While all this is interesting, we might try to have a bit of fun here with what is usually a serious set of topics and create a Crime Paranoia Index to see which state has the greatest gap between the expectations/perceptions of crime and the actual level of reported crime.

For the latter, the ABS has another handy little report called Recorded Crime – Victims, Australia, 2008 which they describe as:

Contains uniform national statistics relating to victims of crime for a selected range of offences that have become known to and recorded by police. The statistics provide indicators of the level and nature of recorded crime as it relates to victims in Australia and measures change over time.

This publication provides a breakdown of the selected offences by: victim characteristics (age and sex); the nature of the incident (weapon use and location); and outcome of police investigations at 30 days. These data are also available by state and territory. The collection also provides for selected state and territories information about the relationship of the offender to victim and the Indigenous status of a victim.

There are five categories of crime for which we have both the ABS perception results and the actual reported offences of crime that we can use here; car theft, other theft, assault, sexual assault and burglary.With burglary, we can use the “Housebreakings/burglaries/theft from homes” for the perceptions data and the “unlawful entry with intent” data from the reported crime data set.

If we look at just the raw data to start with – the estimated number of people concerned about crime types vs. the reported number of offences, this is what we get:


Next up, we’ll create a ratio of perceptions-to-crime by dividing the number of people concerned about a crime type by the number of offences reported for that crime type.

This tells us how many people are concerned per reported crime.


For example, Australia wide there was 43.8 people concerned about car theft for every one actual reported offence of car theft.

Now we can rank these results from highest to lowest. So for each category, the State that gets the highest ratio gets a score of 1, the state with the second highest ratio gets a score of 2 etc etc.


Now we have our results, we can aggregate them to see how all the states are ranked in terms of their crime paranoia.


Victoria is the state which has the greatest gap between crime expectations & perception compared to actual reported offences – being ranked first in car theft and assault, second in sexual assault and burglary and coming in third on other theft.

Compare that to South Australia – the least paranoid state – where they came in last on sexual assault and burglary, second last on assault and other theft, while being 4th on car theft.

Not only did Victoria come in first on the paranoia rankings, but they came in first by a long way – daylight was second.

So, all you paranoid Victorians out there – look out behind you!! 😀