Archive for the ‘ Statistics ’ Category

Examining the factors of Indigenous participation in Crime

Thats the gist of my statistics thesis thats taking up so much of my time right now.

I’m currently working with Anna Ferrante from the UWA Crime Research Centre on a project to examine some Australian Bureau of Statistics data from 2008. Currently reading my way through reams of papers on the subject and so far the answer is a resounding “we don’t quite know.” 

Currently reading my way through reams of papers on the subject and so far the answer is a resounding “we don’t quite know.”

Or more the fact that there is no real easy answer. The combination of social biases, along with substance abuse and socioeconomic inequities all contribute from the base reading of done.

The bad news from this perspective is that at the moment there appears to be no silver bullet for this hot button issue.

However, with the untouched data of over 13,000 anonymised persons at my fingertips I can only hope that this analysis proves to be helpful to this already wide body of research.

And with any luck I will get a chance to carry some of this experience to when I start my Masters in Statistics.

Twitter Sparkline Generator using Unicode

NB: This post uses examples of Unicode that may not show up in some browsers.

One of my main gripes with twitter is the ability to add only text. People often have the desire to share small snippets of data, but to no avail. The ideal idea to share data in such tiny chunks of data Edward Tufte idea of a Sparklines.

For those of you disinclined to read the wikipedia page, sparklines are “data-intense, design-simple, word-sized graphics”, designed to be entered inline with text, at similar height to help illustrate an idea.

Now I am not the first person to suggest entering sparklines in to twitter, in fact the second entry for a google search for sparkline turns up Alex Kerin’s article. However, there are two slight problems with Kerin’s implementation. Firstly, the unicode block characters he is using are not designed to be lined up, and examples that are shown on his page demonstrate this. To be fair, this isn’t his fault at all as unicode compliance isn’t 100%. The second is that a bar and a line can provide two very different perceptions: bar charts generally being used to display discrete data (or continuous data being shown as discrete) and line charts being used to show continuous data – for the record there is no good time to use a pie chart.

To this end I have created a tool for producing two different types of sparkline from an input data source – A crude line graph and a 5-figure box-plot.

Here is an example showing this are using the June 30th 2010 Perth weather data from the Bureau of Meterology, with bars delimiting 3 hour blocks:

The weather yesterday in Perth was quite cool (4.1┣▇▇|▇━━┫17.7) with a maximum of 17.7 degrees occuring around 2pm, before quickly cooling down until 3pm. (⣤⣤⣀⎸⣀⣀⣀⎸⣀⣀⡤⎸⠴⠚⠛⎸⠛⠛⠙⎸⠒⠒⠒⎸⠒⠲⠶⎸⠶⠶⠶).

Limiting this example further, restricting ourselves to the 140 characters of twitter:

Perth 30/06/10: Cool (4.1┣▇▇|▇━━┫17.7), max at 2pm, cooling to around 13°C after 3pm, steady afterwards. (⣤⣤⣀⎸⣀⣀⣀⎸⣀⣀⡤⎸⠴⠚⠛⎸⠛⠛⠙⎸⠒⠒⠒⎸⠒⠲⠶⎸⠶⠶⠶)

This is a 115 character weather report leaving 25 characters for a url to the full data. This may be for temperature only, but it shows the potential and can place 2 dataset in a twitter post with commentary.

I think the boxplots look quite good, however the tool does take a few liberties with the braille layout, relying on people to see a pair of vertical dots as a value in between the two, but it helps convey the message quite well in a limited, text-based format.

Why you are still safer in church than in the strip club.

A recent article from news.com.au has pointed out how much safer Australians are at the local strip club, compared to the church, what is fails to point out is much of the story behind how these figures were calculated and what if any implications the figures have.

The first major issue in the article (aside from the lack of attribution to the data besides the vague “latest data”) is the liberal use of the word “in”:

‘…in the state’s “places of worship” in 2008.’
‘…just as likely to be assaulted or robbed in the sanctity of a church…’
‘using marijuana in places of worship’.

Without the knowing the exact publication to verify the methodology, we will examine the latest publication on crime from the NSW Bureau of Crime Statistics and Research, “The nature of assaults recorded on licensed premises”. This report discusses crimes in or around licensed venues, and breaks data down by the crimes location relative to a licensed venue – weather it was indoors or outdoors on premises, the footpath outside or near or not near to the premises.

“Could you be mugged in the confessional? The answer may surprise you”

If we assume that the Bureau uses similar methodologies in the unreferenced report, we can also assume that crimes ‘in’ churches may refer to crimes merely near or on the grounds of churches as well as inside. Once this emotive language is neutralised (“Could you be mugged in the confessional? The answer may surprise you”) the rest of this article begins to slowly fall apart.

The next big error is the comparison:

A breakdown of the figures showed that 85 people were assaulted in places of worship, compared to 66 at an adult entertainment premises.

Yes, there were more assaults at places of worship compared to adult premises, however it says nothing of the populations these are drawn from. According to the Australian Bureau of Statistics about 53% of NSW Citizens recorded a religion in the most popular choices and has a population of about 6.5 Million people. If we assume a modest 25% of religious people regularly goes to church that gives about 750,000 who regularly attend church, meaning a little over 1 in 10,000 people were assaulted in or around church.

Comparing this with males only, to get a similar population of strip club attendees requires about 23% of males to attend strip clubs with the frequency they attend church to make the comparison made in the article valid.

Before people point out the two issues with my own analysis, yes I am aware females also frequent strip clubs or other adult venues list in the article, but for the purposes of rough estimation we are leaving them out. The other issue is my own numbers assume only religious people attend church, which will be dealt with shortly.

So aside from the fact that 23% of males wouldn’t attend adult venues as often to make the comparison valid, there are still other concerns. Now, lets start looking at why the crimes are being committed. It is not hard to assume that the majority of crimes at adult venues would be influenced by alcohol and drugs or organised crime associated with these venues. furthermore, the bouncer on the door of the strip club may steady peoples emotions a little better than the priest, although neither have quite the ability to keep people in check than the man (or woman) upstairs.

But to look at church as an area of crime is a harder matter, why would someone commit a crime there? Well, ask any teenager and they will have at least had a passing though at drinking or having sex in the most unusual places – churches, graveyards, school grounds, etc… – just to get at the man.

So rather than the father toking up behind that curtain while taking confession, its most likely someone who is there well outside of mass hours
.

Along with this, every Sunday 100s of people show up at church in the morning and leave their cars unattended in quiet suburban car parks while the more secular citizens are fast asleep. This along could account for the number of theft from cars, where as adult venues are usually in nightclub areas and, to their patrons relief, don’t have car parks directly associated with them.

Lastly, many churches offer many services for the underprivileged, and while it is painting a large population with a broad generalisation, the desperation of homelessness can drive people to crime. So while there may be incidents at churches there could be other factors in play to explain them.

In summary this article does little but regurgitate the statistics it has been given and make broad generalizations based on them without trying to understand why they have come about. So, at least for the mean time, its a safer bet to head to the pulpit and drop some coin in the collection plate, than head out on the town with a fistful of singles.