17 July 2009

Gratuitous misuse of statistics to kick Kambah

| johnboy
Join the conversation
18

A Canberra Times’ sub-editor has taken Kambah bashing to a new level with this headline:

    Drink-drive alarm: Kambah motorists caught out

It’s peculiar because their source, analysis from the NRMA-ACT Road Safety Trust, lists Civic as the leading sources of drink drivers with 97 convicted compared to 53 in Kambah.

But it gets downright pernicious when you consider that Kambah, being a geographically vast suburb, is Canberra’s most populous.

According to population.canberra.net.au The population of Kambah, at over 16,000 is far above the next closest suburbs. Braddon, third in the drink drive stakes has just 4,600. The number one place holder for drink driving, Civic/City, has just 2,450 people.

So Kambah residents are in fact being extremely well behaved. Doubly so when you consider that Braddon and Civic residents have many pubs within easy stumbling distance of home.

Join the conversation

18
All Comments
  • All Comments
  • Website Comments
LatestOldest

All figures for Kambah should be divided by 4. Same for Wanniassa.

I’m concerned at the 7 people in the report dealt with by the Magistrates court for whom gender was unknown (Table 1). Not identified under the GILBT, gender unknown!

Does anyone know anything about this Smithworks Consulting? They appear to be unfamiliar with sophisticated statistics, analysis or presentation. What appears to be their website (http://thesmithworks.com/) is unusually uninformative.

Thoroughly Smashed1:42 pm 17 Jul 09

54-11 said :

Several years ago I had the “pleasure” of an interview with a CT reporter, which took about 30 minutes. Just a few sentences were reported, which completely missed the points I was making and gave almost an opposite view of what I was saying.

I will never talk to a reporter again.

I also know someone who managed to attract strong criticism from numerous CT letter writers for something they never said to the CT reporter that interviewed them. The offending words under were essentially fabricated, and I understand the CT reporter ended up being cut off by their organisation’s media department.

“So Kambah residents are in fact being extremely well behaved” err 53 of them (caught) were not.

…. I’ll take more care driving through Kambah 😉

Thanks for the link Caf.

I don’t believe the journalist has even looked at the report.

“…Chief Minister Jon Hargreaves’ proposal to name and shame drink-drivers…” is probably the single most terrifying phrase I have read in the CT for some time.

PS: http://the-riotact.com/?p=2706
See you at the Christmas Party, John!

(As should be clear from my other post: the one-in-seven Civic residents bit is complete bulldust, and the journalist responsible should have spent 20 minutes reading the report to find this out).

The original report is available here.

The figure quoted in the article of 97 convicted drink drivers for Civic is actually arrest location, not residence (the report analysis both). In reality only 1 single Civic resident was convicted of drink driving. Braddon was certainly not third, either, with 16. Kambah was the highest single suburb, but as noted it is also the most populous so this is to be expected.

In terms of a per-capita basis, the worst suburb was actually Pialligo, where 4 of their meagre 113 residents were busted. Of the suburbs with decent sized population, Gilmore had by far the highest rate of drink-driving.

Perhaps most important is this conclusion (referring to residence of offenders):

There is some variation between suburbs, but remarkable uniformity across the Canberra regions of Gungahlin, Belconnen, North and South Canberra, Woden, Weston, Tuggeranong and Gordon.

(Why has the report’s author elevated Gordon to region status? I have no idea…)

fnaah, I guess the problem was not what Aitken said, but how it was reported.

Several years ago I had the “pleasure” of an interview with a CT reporter, which took about 30 minutes. Just a few sentences were reported, which completely missed the points I was making and gave almost an opposite view of what I was saying.

I will never talk to a reporter again.

Brilliant bit of thinking this is, one paragraph directly contradicts its predecessor:

“The figures show nearly one in seven Civic residents appeared on drink-driving charges.

Professor Aitken said the statistics revealed recidivism was a large problem in drink-driving.”

You guys are terrible! The Canberra Times would never display poor journalism. Silly talk.

The same article has also promoted John Hargreaves to Chief Minister. Do they know something we dont?

It’s just another example of a reporter (who should know better) taking raw statistics and fail to do any sort of analysis.

The point made by the OP is a valid one – the first bit of analysis should have been to do a comparison based on overall suburb population. The next would be to look at longer term trends, and their critique based on that.

As Wraith says, CT at its best.

Canberra Times reporting at its best again…………….

Postalgeek said :

Not so sure about comparing Braddon and Civic with Kambah, given that many people who drink in Civic and Braddon don’t actually live there and are caught en route to their home turf.

Not that it is what the report is based on anyway, but a lot of other Tuggeranong drink drivers would be caught in Kambah as well, as they usually have the booze bus on the parkway.

2902 SPEW! that’s my new tattoo 🙂

The report was on residents.
So the Civic figures were for residents of Civic, not people caught there.

Not so sure about comparing Braddon and Civic with Kambah, given that many people who drink in Civic and Braddon don’t actually live there and are caught en route to their home turf.

Daily Digest

Want the best Canberra news delivered daily? Every day we package the most popular Riotact stories and send them straight to your inbox. Sign-up now for trusted local news that will never be behind a paywall.

By submitting your email address you are agreeing to Region Group's terms and conditions and privacy policy.