Beyond The Commons

Aaron Wherry covers all the goings-on in and around Parliament Hill. Follow Aaron on Twitter: @aaronwherry

Categories: Beyond The Commons, Capital Read

The chief statistician steps aside

by Aaron Wherry on Wednesday, July 21, 2010 7:45pm - 0 Comments

Munir Sheikh has posted a statement on the Statistics Canada website.

I want to take this opportunity to comment on a technical statistical issue which has become the subject of media discussion. This relates to the question of whether a voluntary survey can become a substitute for a mandatory census.

It can not.

Under the circumstances, I have tendered my resignation to the Prime Minister.

Munir Sheikh had been a member of the public service since 1976 and was named chief statistician by Prime Minister Harper in February 2008. A statement from the office of Industry Minister Tony Clement after the jump.

I acknowledge with regret the resignation of Munir Shiekh, the Chief Statistician of Canada.

There has been considerable commentary about the federal government’s decision to replace the 2011 mandatory census long form with the voluntary National Household Survey.

The Government took this decision because we do not believe Canadians should be forced, under threat of fines, jail, or both, to divulge extensive private and personal information. We believe it is not appropriate to compel citizens to divulge how many bedrooms they have in their houses, or what time they leave for work in the morning. The Government’s approach is about finding a better balance between collecting necessary data and protecting the privacy rights of Canadians.

As I have noted previously, Statistics Canada’s preferred approach would have been to maintain the mandatory long form census.

However, after the Government’s decision to replace the mandatory long form census Statistics Canada was asked to provide options for conducting a voluntary survey of households. One of the options provided - the voluntary National Household Survey – was chosen.

A voluntary long form survey offers challenges that do not exist in the case of a census that uses coercion to compel completion. Nonetheless, by working together with the professionals at Statistics Canada I believe we can compensate for these challenges and offer data-users high quality and accurate information.

I have relied throughout this process on the frank and open advice of Statistics Canada and the Chief Statistician. I would like to take this opportunity to thank all employees of Statistics Canada for the hard work and dedication that has made Statistics Canada one of the best national statistical organizations in the world.

Until a permanent successor can be found Wayne Smith, Assistant Chief Statistician, Business and Trade Statistics, will act on an interim basis.

Tags: census, Munir Sheikh, Statistics Canada, Tony Clement

More by Aaron Wherry

Out There

All of this because Stephen Harper is seemingly incapable of admitting error, changing his mind, or ever backing down. And his ministers have turned, effectively, into robots who do and say what they're told. How horrible it is that our country has been captured by this bunch.
Ariadne

How long is long exactly, to make data more valid and accurate? How would Can Stat know whether somebody lies or fibs on mandatory census? Has anybody been penalized for lying on Census? Is there a way to check the veracity of each answers? If not, then how can they say that datas in mandatory census is more reliable than voluntary? If not, does it mean they already know the answer to those questions and should not have asked those questions in the first place?
Ariadne

How long is long exactly, to make data more valid and accurate? How would Can Stat know whether somebody lies or fibs on mandatory census? Has anybody been penalized for lying on Census? Is there a way to check the veracity of each answers? If not, then how can they say that datas in mandatory census is more reliable than voluntary? If they have the means to check them, does it mean they already know the answers to those questions and should not have asked those questions in the first place?
http://intensedebate.com/people/RobinBC RobinBC

Who decided what questions would be put into the long census form and what were the reasons.? What authority did those who put the questions on the form have to do so?

Derek

In general, census questions are largely the same over time (this is required to ensure comparability of results from one census to the next). Small changes (additional questions, changes in wording, etc.) may be required periodically to ensure the questions reflect current socio-economic developments and policy requirements. Such changes are generally proposed by StatsCan staffers, go through a public consultation process and are finally tested in the field to ensure the wording will ensure accurate results.

Finally, the propsed questions are brought before cabinet, who has the final authority. The results are then published in the Canada Gazette.

This, at least, is my understanding. It appears the usual protocol was not followed this time around and Cabinet chose to simply impose its authority.

http://intensedebate.com/people/RobinBC RobinBC

A look at the questions on the long form merely confirms my belief the Government was correct in making them voluntary.
"Tested in the field to ensure the wording will ensure accurate results."? Is this voluntary? How could they then be accurate?

http://intensedebate.com/people/Jenn_ Jenn_

Incidentally, while Sheikh couldn't say what he really felt about Clement, he wasn't so hampered in what he really felt about Ivan Fellegi, huh?

In closing, I wish the best to my successor. I promise not to comment on how he/she should do the job.

I love the brevity and power of this guy's writing.

http://intensedebate.com/people/Halo_Override Halo_Override

Comparing the density and meaning of his writing against what we saw from Mr Latulippe this week is certainly instructive on the nature of personal character and how we express it in what we do.
Lord Kitchener's Own

I took that slightly differently actually.

I read that not as "I wish Ivan Fellegi had kept his mouth shut" but as "I understand whoever they get to replace me has probably got kids to feed, so even though I'm sure that person is going to think Clement is an idiot too, I promise to stay quiet and not make his or her life more difficult than it has to be, as it's going to be hard enough for the new Chief Statistician to swallow what they're going to have to swallow without me chiming in from the sidelines on top of it all".

NorthernPoV

Can't read 'em all – too many cranks on here and I gotta go to work.

I DO find the latest meme hilarious, though

- slagging the stats guy for his timing: "why didn't he resign weeks ago?"

It does expose the absolute paucity of the gov't position here.
Poker Face

I think that this was entirely Clement's thing – thus the PMO's odd silence on such a polarizing issue. I feel like the next ministerial shuffle will involve Clement's card moving toward the bottom of the deck.
The appropriate option would be to reform/amend/shorten the long form, or lengthen the short one (only by 4-5 questions) while ditching the long form.

(I carry a CPC card, for what it's worth. Never been a big fan of Clement in particular though.)
maudie

I read that Clement tried to advise Harper against this dumb move, but was overruled. Harper is the worst "leader" we've ever had in Canada, bar none. He makes even Brian Mulroney look good. (Well, maybe that 's going too far; let 's just say, "better by comparison.") Please, people, can we have the Liberals back for a while to set this sorry house in order again?
Guest

Thank you Mr. Sheikh, for keeping your head, and keeping true to universal righteous principles under extreme adversity. You have done a great service to Canadians everywhere. I do not trust any successor to your job unless Harper backs down or they keep resigning.