Standard Error

The target of one of Tim Blair's five-minute hates on Wednesday was fellow journalist Peter FitzSimons. FitzSimons' crime? He said there had been an estimated 100,000 civilian deaths in Iraq. Now you could quibble that the Lancet study measured all excess deaths rather than just civilian ones though since the vast majority were civilian deaths, this isn't that big a deal. Blair, however, calls FitzSimon's "stupid" and links to Andrew Bolt's train-wreck of a critique. I suppose that Bolt's isn't the worst of the critiques out there, but with Fumento around the competition is very very stiff.

Blair writes:

FitzSimons is famously sensitive to criticism, which may be why his non-sport columns aren't linkable.

There may be a similar explanation for the fact that critical comments are banned from Blair's blog.

Blair continues:

And paint, to cover Fitzy's house with charming peace messages.

If I had some paint, I'd write this formula on Blair's house:

s/√ n

That's the formula for the standard deviation of a mean. Notice the divided by √ n part? That means that as n, the sample size, gets bigger, the standard deviation and the sampling error gets smaller. Unfortunately Bolt just doesn't understand this because he writes:

But the survey techniques they used to give clearly wrong figures in Fallujah are the same ones they used in the other 32 clusters of households that they interviewed elsewhere in Iraq.

Don't worry kids! If you fail maths at high school, you can always get a job writing for the Herald-Sun or the Bulletin.

More like this

The latest pundit to have a go at the Lancet study is Andrew Bolt. Like most of the critics, Bolt just does not have the statistical background to produce a competent critique. In Bolt's case this is even less excusable, since he had the benefit of the Economist's excellent…
One of the few things that Andrew Bolt got correct in his original criticism of the Lancet study was the sample size, 988 households: Its researchers interviewed 7868 Iraqis in 988 households in 33 neighbourhoods around Iraq, allegedly chosen randomly, and asked who in the house had died in the 14…
One of the many factual errors in Parkinson's piece on deaths in Iraq was the claim that the Lancet study only surveyed 788 households (actually it was 988 households). I did a Google search to see who else had made the same error, and what do you know, it first appeared in a error-filled May 25…
Via Suki Lombard I discover that the Australian government's position on Iraqi deaths because of the war is that the Lancet estimate of roughly 100,000 excess deaths is an exaggeration and we have no idea how many have died and no plans to find out. Govt seeking no information on…

Standard errors, changing standards, standard nonsense. It's called blogging lite, just another day in the oxygen free suburb of Spleenville. Quoting or linking to Bolt as a factual reference? He might as well refer to a Japanese schoolgirls diary.

It's a tough order, but I think I've got the link for you: 'Simple Life' Denies Deer-Poaching Claim:

Paris Hilton's reality show "The Simple Life" is disputing allegations that a deer was illegally killed so that it could be gutted by the socialite and her sidekick, Nicole Richie.State officials said the deer was killed Tuesday night and they believe it was to be gutted for the show. Investigators are looking into accounts the deer was shot by someone firing a gun from inside a car.

And since Tim Blair can't help himself in giving a free kick to animal activists (even if these ones have badges), guess which side he'll take. Just like every other thread on Spleenville, it's a reprise of that old Kenny Everett film "Godzilla versus Bambi" - and you KNOW who wants to be Godzilla...