Je dédie ce billet à l’arrogant et imbuvable Alain Dubuc.
Your data or your freedom
By George Jonas
Using my first computer in 1986, I suggested it was insolent for census bureaucrats to threaten Canadians with jail unless they revealed the number of bedrooms in their homes (and within four days, at that). Every 10 years or so, I re-visited the topic, arguing that for social engineers to hitch a ride on the compulsory short census was intrusive and unnecessary.
Twenty-four years later, a government agreed: This summer, Industry Minister Tony Clement instructed StatsCan to scrap the mandatory long form and replace it with a voluntary one. Social engineers could have their data; they just couldn’t expropriate it.
The reaction was enlightening. Mauled by their natural ally, government, the elites of statism — planners, bureaucrats, regulators — argued for more government. They argued that Canadians demanded coercion. Yes, people wanted to be asked intrusive questions … but, no, they wouldn’t answer them unless they had to.
Talk about tour de force.
Marc Garneau, the Liberal critic for Industry, Science and Technology, gave us a glimpse of the mindset in these pages. “If a stranger stopped me in the street and asked me how many bedrooms I had in my house, I would tell him to get lost,” he wrote. “When the government of Canada asks me the same question in an official census long-form questionnaire, I react in a completely different manner.”
For the Liberal member from Westmount-Ville-Marie, officialdom transformed the impertinent into the pertinent. Alchemists of statism saw a stamped paper or badge as a magic formula that turned the base metal of intrusion into the gold of civic duty.
“Thank goodness most Canadians agree with me,” Mr. Garneau wrote. “Thank goodness most Canadians aren’t paranoid and fearful, believing that they owe nothing to their country. Thank goodness that most Canadians believe in the notion of civic duty.”
But if so, why do they have to be threatened with jail? Might they see their civic duty differently from Mr. Garneau?
Might Canadians regard it their civic duty to deal with questionnaires according to their own lights and conscience, as free people in a free society? Might they see it as the government sees it, for once, amazingly, to the complete consternation of the statist elites?
I don’t know. Neither does anyone else.
The dispute isn’t about statistics but coercion. An offer of un-coerced data to a statist is like an offer of consensual sex to Jack the Ripper. Statists are rapists. They don’t know how to deal with “yes.” How are we to plan with voluntary information, they sputter. How are we to social engineer?
Hmm. Entrepreneurs who can’t compel data plan at least as well as governments that can. Does it hamper business that it can’t coerce? Perhaps it would be salutary if the state’s planners and engineers did less planning and engineering, but it’s not going to happen. The state will poll, research, extrapolate, solicit and buy data, like everyone else.
Statists blow their stack at this. Only a Neanderthal, a statistical illiterate, can argue for polls or extrapolation taking the place of the mandatory long form.
I turn to a statistician. “Don’t quote me,” he replies, “but we send the mandatory long form census to only 20% of the households. We don’t use the word ‘extrapolate,’ of course, but…”
Munir Sheikh, StatsCan’s former captain who went down with his mandatory ship, told a CBC interviewer this week that the response rate to the old long form, after follow-up calls, was 100%. Considering that the coercive exercises of Romania’s notorious dictator, Nicolae Ceausescu, never exceeded 99%, StatsCan’s performance was impressive. Apparently the Harper-government’s estimate of voluntary compliance is a mere 50%.
Interesting. The mandatory long form went to 20% of Canadian households. There’s no reason for the voluntary form not to go to every household, is there? While 100% is more than 50%, 100% of 20% is less than 50% of 100%. And even if the voluntary form went only to every second household, 50% of 50% would still be more than 100% of 20%.
I wish the CBC had asked Mr. Sheikh. He might have had a good answer. Or not. The Corp wanted to nail the Tories and was taking no chances.
Had he been asked, Mr. Sheikh’s might have replied that whatever the numbers, voluntary responders aren’t as valuable as coerced responders from the statistician’s point of view. 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.
What a waste of ink on a trifling matter, some say. Happy Canada! In a world awash with misery and oppression, Canadians worry about whether or not to reveal the number of bedrooms in their homes to a census taker.
I suggest if we aren’t awash in misery and oppression, one reason is we worry about trifling matters like liberty. When we stop worrying, we’ll be on our way.