Observant visitors to this blog may have noticed the recent appearance of a “wiki” button at the top of the page. This links to the recently established Stubborn Mule wiki, which I plan to use as a repository of information relevant in some way to the blog. Since so many of the posts here focus on data analysis, I have started with a collection of links to useful sources of data online, particularly economics and finance data.
The latest link I have added is to the New South Wales Bureau of Crime Statistics & Research (while I did not include it in the economics and finance section, maybe it does belong there). This site includes a research data set which provides monthly crime data going back to 1995 broken down by local government (council) area and offence type. Needless to say, the first thing I was interested to learn was the level of criminality in my own local area, particularly as I moved here only very recently.
The chart below shows the total number of crimes in the various offence categories for 2008 in my local government area of Marrickville. While I was not surprised to see theft coming in at the top of the list, there were a few oddities further down. I was initially surprised to see driving offences at the bottom of the list. My driving is, of course, impeccable but I do not know if the same is true of all of my neighbours, not to mention visitors to the area. Digging further, I discovered that from 2003 onwards*, the figures for driving offences have been zero for all areas and transport regulatory offences have leapt up. So, presumably there has been a classification change. One mystery solved.
Marrickville Crime Count (2008)
More intriguing is blackmail and extortion. Until 2008, the highest rate this crime had reached in Marrickville was four cases per year and in three years, the figure was zero. Yet, in 2008, there were nine cases of blackmail and extortion. What lies behind this wave of blackmail around the corner? Mystery not solved.
This led me to examine other trends through time. Starting with theft, I was gratified to learn that 2008 was the lowest year for theft since these records began. I am hoping 2009 will be lower still.
Occurrences of Theft in Marrickille
A look at prostitution also suggests the area has become more law-abiding after a significant spike in offences in 2001.
Occurrences of Prostitution Offences in Marrickville
As for serious crime, Marrickville experienced three homicides in 2008. The total number of homicides in the area since 1995 is 66, putting Marrickville in a somewhat disturbing 14th place out of 155 local government areas, although these two have been reducing over recent years. For those interested in the most murderous areas in New South Wales, here is a list of the top five areas in terms of total homicides since 1995. Any country readers will note that all of these local government areas are in Sydney (the area in the table labelled “Sydney” encompasses only the central business district and some inner-city suburbs).
| Area | Homicides |
|---|---|
| Fairfield | 242 |
| Sydney | 327 |
| Blacktown | 136 |
| Liverpool | 102 |
| Parramatta | 82 |
* The historical data for Marrickville is in the “Files” section of the blog.
Possibly Related Posts (automatically generated):
- COVID-19 by Suburb in New South Wales (2 May 2020)
- Australian Grocery Prices on the Rise? (1 September 2008)
- Benford’s Law (16 April 2012)
- Where Have the Fish Come From? (11 August 2009)
I guess Newtown crime is distributed between Marrickville and Sydney then.
LGA seems a fairly coarse grain – I wonder why it wasn’t done by suburb ?
Surely you need per-capita figures for comparisons between suburbs to be useful?
As a resident of 2204 for 10 years I can attest to a distinct demographic change over that time. Using my precise measurement technique ‘Suits at the Bus stop’ there was a steady gentrification over the first 8 years that has notably accelerated over the last 2 years. Perhaps this may explain some of the decline in thefts offences but I think that the rate of theft offences state wide has generally declined with the benign general economic conditions. The figures since the start of the GFC are keenly awaited!.
Mayor: Good question. They would clearly have the more granular data, but perhaps there are some privacy concerns.
Danny: I must admit that the primary reason for not calculating per capita rates is that I have not yet found population data by local government area. If you find it, let me know! Mind you, it did also occur to me that, from the perspective of the risk of personal exposure to crime, perhaps crime rate per square meter would be more relevant than per capita.
Paul: you are in good company in raising concerns about the impact of the GFC on crime rates. We shall see.
S, you might like to look at the comparable figures for London http://maps.met.police.uk/ they get broken down to ” sub-ward” the smallest policing area. T
Sean: You should be able to get population figures all the way down to local census collection areas from the ABS. Somewhere. Somehow. They’re core to Antony Green’s modelling of elections.
Something which hasn’t been mentioned yet is that the figures being tracked are offences, which are not just subject to the prevalence of the crime in question but the policies of law enforcement — especially if we’re talking about victimless crimes like prostitution and drig offences. I daresay that 2001 spike in prostitution offences was the direct result of a specific police campaign to enforce the laws on such things — and that sort of thing is often driven in turn by a political imperative.
Sean,
This is a general comment on “The Stubborn Mule”, a blog I enjoy reading from far away.
I can see that you enjoy measuring things, which I find interesting. However I end with a feeling of lack of conclusions. I have hope that measurement may lead to some concentrated conclusions.
Bob
Bob: It is a fair point you make. I would say that part of the problem is that in my measuring, I often stray outside areas of my expertise and so tend to be tentative in drawing conclusions, preferring the comfort of knowing how to measure and display the results. Whether or not I am on safe ground, I should be bolder in drawing conclusions, particularly since this is likely to elicit more comments and all bloggers crave comments. (So thanks for yours!).