Monday, August 16, 2010

Because There Is Such A Thing As Truth

George Jonas' latest on the census long-form plumbs the depths of innumeracy:

At least, that’s the mantra. Self-selection skews data. No one explains why coercion doesn’t skew data the other way, or why doesn’t it bother statisticians if it does, or why data can’t be adjusted for self-selection when it can for most other things.

The answer to the last bit, about adjusting for self-selection, is of course that data can be adjusted for self-selection where you have census data to correct it with.

But its the first part that's most interesting. Jonas is suggesting, obliquely, that there is nothing but skew--skew this way, skew that way, the further implication being that these directions are "left" and "right", politically, and that statisticians tend towards socialism.

But of course that's not accurate: to any question about the demographics of an area, there is a correct answer, and you get that answer when you have managed to extract (via the mandatory long-form) a truly random sample from within that area (ie a sample that is unskewed). Underlying Jonas' argument, then, is an appeal to truth relativism.

Which, you would think, is weird coming from a Conservative, but perhaps more common than you might think.

8 comments:

Robert G. Harvie, Q.C. said...

So, then, you would accede that, in fact, there is a functioning Church of the Jedi in Canada with 20,000 members (according to the last census)?

Gayle said...

Oh dear lord Harvie.

If some Schmo came up to you and started telling you what the Charter really says, and what the latest SCC ruling really means, would you incorporate that into your practice, or would you rely on your own education and experience?

In other words, the people who study statistics know what they are talking about. It would be wise for you to accept their expertise.

Unless, of course, you think all those non-lawyers showing up in court to defend themselves or their friends are doing a super great wonderful job of it.

Robert G. Harvie, Q.C. said...

Gayle.

To be brutally honest, I have as much faith in reading chicken bones as I have predicting court outcomes based upon what the Supreme Court of Canada says.. but that's a whole other argument.

My point isn't that we must assume that Jedi is a serious religion, it's that for the same reason that we get results that are different when a census is voluntary, we may get results that are different (ie not purely accurate) when they are manditory.

We seem to have no problem making broad social policy decisions based upon little more than superstition (gun registry, more jails), so I think the whole thing is overblown hype all the way around.

Hoever, personally, I think the questions are intrusive and I don't think in a free society that I have any obligation to explain my personal life in any detail unless I happen to ask the government for specific assistance to which the questions relate.

So.

My census, as I've blogged.

I'm a black African Jedi.

bigcitylib said...

"My point isn't that we must assume that Jedi is a serious religion, it's that for the same reason that we get results that are different when a census is voluntary, we may get results that are different (ie not purely accurate) when they are manditory."

No: more accurate. Because mathematicians say so. Are you a mathematics denier now too? Because this is where this line of reasoning takes you (its where it takes Jonas, with his implication that statisticians must be a little lefty). And that's the road to madness.

Robert G. Harvie, Q.C. said...

I'm not saying they are a "little lefty" at all.

I'm saying that any survey, even when it is manditory, is never 100%accurate.

And the "Jedi" response confirms that fact.

Surveys of any kind are presumed accurate based upon statistical theory to a certain percentage.

Making it "manditory" may result in a higher level of response from a broader spectrum - and hence more likely to be accurate - but it hardly guarantees accuracy.

And beyond that, my point - is the questions are getting more stupid as society evolves.

Have you ever met a "black" person?

Not me. I've met tan people, brown people, dark brown people.. and I've also never met a "white" person.

These sorts of questions are the fodder for stupid government social engineering - that we think is all so clever - but, for my dime, will eventully lead us to a fall.

And, anyway, I just think my religion is not Stephen Harper or Michael Ignatieff's business.

So.

This black Jedi says screw the census.

bigcitylib said...

"I'm saying that any survey, even when it is manditory, is never 100%accurate."

The answer to this is duh. You could use the same argument against the whole notion of doing a census in the first place. Given that we want to do a census then that's hardly an argument against the need to draw a truly random sample from a partiucular geographic area. And its hardly an argument that THERE ISN'T such thing as demogrphic truthes about such an area--which is what Jonas implies--that cannot be best attained by getting a truly random sample via a mandatory long form.

Gayle said...

I get your point. MY point is that the experts disagree with you.

Tof KW said...

Gayle said...I get your point. MY point is that the experts disagree with you.

Mr Harvie's stance pretty much sums up the Harper government - simply that they know better than the experts.

From tax reduction (they knew better than the real economists who said cutting the GST was a bad idea), to the gun registry (they know better than the chiefs of police), to tough on crime (they know better than the experts that building more jails will reduce crime), to the environment (they know better than 98% of the scientific community regarding greenhouse gas emissions), and finally now they now know more than the actual statisticians and mathematicians do.

Golly, the Harper government is so smart ...they know better than anyone about anything.