Dispatches From the Moderate Left

Saturday, May 27, 2006

Lies, Damn Lies and Gun Statistics

I don't know what it is with pro-gun lobbys but they have an almost pathological proclivity towards lying through statistics. Take their claim that hunting is safer than table-tennis:
One of hunting's dirty little secrets has been revealed: The "safety statistics" kept and promoted by various government agencies--which purport to show just how safe hunting really is--are a total joke.

Consider, for instance, the Texas Parks and Wildlife Department. On the "Safety First" page of its online hunter-education course, the department declares that "hunting is one of the safest outdoor activities you can enjoy" and backs up its claim by citing National Safety Council statistics that purport to show that hunting is not only safer than fishing and swimming; it's safer than football, basketball, and baseball! Or take this 2004 press release from Minnesota's Department of Natural Resources, which proclaims, "Based on the number of people seeking emergency-room treatment for sports injuries, The National Safety Council reports that hunting has fewer injuries per 100,000 people participating than football, baseball, cycling, volleyball, swimming, golf, tennis, fishing, bowling, badminton, billiards and ping-pong."

Now, you know that this claim is absurd and the way statistics are abused to make the statements factually correct are laughable. All injuries are treated equally seriously ("Getting hit by 200 pellets of birdshot is treated just like spraining a pinkie in ping-pong") and fatalities aren't counted by the parks department at all. Also, the time spent in each activity, which is a good measure of its risk, isn't taken into account. While people may spend dozens of hours a week training and playing sports like football, many hunters who make up the "per 100,000" part of the raw statistics only participate for a few hours a year.

When I posted on Australian gun statistics a couple of weeks ago I linked the dissection of one gun group's distortion of statistics related to Australia's gun buy back. There's so many fallacies in the piece that I won't list them all, but it uses irrelevant stastics (eg. absolute not real growth figures) statistically irrelevant increases, flat out wrong figures and weasel words like "dramatic" where the figures don't actually look all that bad.

An anonymous commentator, trying to convince me that claims of a reduction in gun violence in Australia following the gun buy-back were incorrect, inadvertently displayed another example of the gun lobby distorting Australian gun statistics. The piece is very high minded, trying to portray itself as a tool to educate the public on statistics so that they don't get deceived by statistical trickery:
The mere mention of statistics can be enough to provoke slumber, but statistics are used every day to make points, support arguments or disprove theories.

The problem is that statistics can be misused misunderstood and misinterpreted.

Why yes, yes they can. For their own inscruitable reasons, the "International Coalition for Women in Shooting and Hunting" which put out the paper decide to make a point about they way rate of change statistics could be misused by inappropriately taking into account the statistical blip caused by the port arthur massacre. They note that if you look at the rate of change in the decline in gun deaths before and after 1996 there is no real difference, as long as you don't include 1996 in the second period:



I'm not sure why they picked the rate of change statistic. If I were making a hypothesis about a massive gun buy back I would suppose that it would have the effect of reducing the absolute number of gun deaths and, perhaps, help maintain a long term decline in overall gun violence. There's no particular reason to think that it would accelerate that long term decline. Unfortunately, the Age piece I linked to did use the language of an increase in the rate of decline in gun deaths:
So, 10 years on, can we see a difference? Resoundingly, yes. In the decade up to and including Port Arthur, Australia experienced 11 mass shootings. In these 11 events alone, 100 people were shot dead and another 52 wounded. In the 10 years since 1996 and the new gun laws, not one mass shooting has occurred in Australia. For this reason alone, Australia is a safer place. (In 2002, a gunman killed two and wounded four at Monash University. Five victims are internationally recognised as a mass shooting.)
...
Even before Port Arthur, gun-related deaths, suicides, homicides and unintentional shootings were slowly dropping. But after the tragedy, the rate of decline accelerated markedly. From 1979 to 1996, 11,110 Australians died by gunshot - an annual average of 617. In the seven years after new gun laws were announced (1997 to 2003), the yearly average almost halved, to 331.


But the figures that are used there are averages of absolute numbers, not rate of change statistics. Looking at the raw numbers, not just the averages, the number of gun deaths did have an unusually large fall after the introduction of the gun laws, and this isn't just a decline caused by comparing 1996-1997:

(source)

In case you were wondering, the fall correlated with the gun buy-back is statistically siginificant (1996 was excluded from that analysis so the massacre didn't distort the statistics). Incidentally, that latter paper also highlighted an important point. You can't look at the 1997 gun-buy back in isolation. Australia and Victoria in particular have had a history of tightening gun ownership laws, particuarly in the late 1980s after the Hoddle street and other massacres. Each of these phases of tightening legislation is associated with a statistically significant decline in the number of gun-related deaths and the cumulative effect of these gradually increasing restrictions has probably contributed to the long run fall in firearms related deaths.

Despite the many, many attempted distortions of pro-gun lobby groups (their fingerprints are all over the wikipedia article on gun politics in Australia), the basic point I was making in my previous post still stands. Australia is safer because of Howard's gun buy-back. I can only think that the pathological tendancy of gun groups to distort statistics for their own ends stems from a refusal to admit that the real world cost of widespread gun ownership in society is measured in blood. I can understand why they're uncomfortable with that, but it doesn't mean we should listen to them.

4 Comments:

  • Good morning Jeremy!

    Did you know you were on Crikey Blogwatch yesterday?

    Hope tax law is treating you well.

    By Anonymous Laura (catonthebench), at 11:58 AM  

  • Hey laura. I did notice a few in-clicks from Crikey, it seems a bit strange that they linked such a US-centric post, but hey :).

    And tax started treating me a whole lot better when Ann excluded about half the syllabus from the exam! Good luck with it...

    By Blogger Jeremy, at 12:09 PM  

  • If you're a moderate left. Then I'm a Monkey's uncle. You're just another ranting and complaining, whining, spineless liberal loon. Believing you know what's best for everyone else, and the rest of the world better fall in line. Well sorry, I have a brain, and I use it. I have a right to bear arms, and that's what I'm going to do.

    At the end of the day, I would rather have a gun to protect myself than have to wait 15-20 minutes for the police to show up. That simple.

    By Anonymous Anonymous, at 3:44 PM  

  • What are your thoughts on this then?

    http://www.law.harvard.edu/students/orgs/jlpp/Vol30_No2_KatesMauseronline.pdf

    By Anonymous Anonymous, at 9:41 PM  

Post a Comment

<< Home


 

Listed on BlogShares