- StatsCan Performs Too Important a Roll in Canada to be Corrupted by Harper

Posted: 1:20 PM on February 13, 2011 The Globe and Mail
Chief statistician asked to rethink census for 2016
Steven Chase and Tavia Grant, The Globe and Mail, Feb. 11, 2011
http://www.theglobeandmail.com/news/politics/chief-statistician-asked-to-rethink-census-for-2016/article1904832/

"When it comes to the 2011 census, however, Mr. Smith said he is sure about one thing: there’s no justification for critics to say that moving to a voluntary long-form survey will wreck the quality of the data.

'I’m asking Canadians to suspend judgment because there’s no scientific basis for saying this is going to be fundamentally flawed.'


'There is no scientific reason why you would say before it even starts, before I see results, that there’s going to necessarily be a significant problem with the count of Inuit or Métis or immigrants beyond the levels we’ve seen in the 2006 census.'"

It is quite surprising that the head of StatsCan would put such an assertion in absolute terms - vis.:

"no scientific basis for saying this is going to be fundamentally flawed"; and,

"There is no scientific reason why you would say before it even starts . . . going to necessarily be a significant problem".

The statements attributed to Wayne Smith in this article are the type of statement, highly partisan and political, you would expect from Tony Clement or Stephen Harper in an effort to obscure and obfuscate the real underlying issues - not the least of which is political interference and loss of trust in Stascana and the result of the Census.

I would not expect them from a scientifically trained professional, and certainly not from the Chief Statistician of StatsCan.

When you consider the expense in doing the Census, their importance to good government and the private sector, a risk-benefit analysis is in order - vis.:

'What are the chances that the change made to the Long From Census may result in less accurate information. Then, is it worth this risk.'

This is the very nature of what Statisticians do and are trained for. This is precisely the type of thing StatsCan is there for.


An applicable absolutism is:

StatsCan ought not to make such changes if they are unable to answer, and in advance, precisely the issue Smith has so extremized - i.e. 'what are the chances the results will be fundamentally flawed, a significant problem'

Otherwise they, or should I more accurately state, Harper, Clement and the Con's are literally rolling the dice with Canada's future.

It seems to me that any self respecting, objective, disinterested scientist would understand this.

If StatsCan is unable to make such evaluations, under Wayne Smith, then he simply ought to hand over the reigns to someone who can 'make it so'.

Munir Sheikh, Canada’s former chief statistician at StatsCan, wrote, when he resigned, a very insightful article regarding the problems with the changes enforced by Clement and Harper.

They raise an issue of the quality of the results - i.e.

- " Quality means that the data Statistics Canada releases must accurately reflect the phenomena it is trying to capture"

- "The other element of quality is trust in Statistics Canada’s numbers. "

"By making a decision on a technical issue – which the government has every right to do under the current legislation – the government risks the creation of a trust gap.

It takes ages to establish credibility. It takes much less to tarnish it.

The issue of Statistics Canada’s independence is a serious matter,"
(Munir Sheikh, G&M, 10 Aug.'10)

StatsCan performs too important a roll in Canada to be corrupted by Harper and the Con's.

They have already done a lot of damage.

All Canadians should do a cost -benefit analysis of the Harper reign and I am confident if they do Harper will get the boot.

excerpt: Lloyd MacILquham cicblog.com/comments.html