On February 01, Philip Cross, Chief Economic Adviser at Statistics Canada announced his leaving the agency. He follows the head of the agency, Munir Sheikh, who resigned last year over Government plans to redesign the Census. Mr. Cross is leaving for much the same reason.
At issue is replacing a compulsory census questionnaire with a voluntary questionnaire. In essence, this means replacing a random sample with a discretionary sample. Discretionary samples have their applications, but making inferences about the population isn’t one of them. Unfortunately, making just these types of inferences is whole point of the Census. That’s why Mr. Sheikh wrote in an open letter to the Government:
“I want to take this opportunity to comment on a technical statistical issue which has become the subject of media discussion … 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.”
Mr. Cross is quoted in the Globe and Mail as expressing concern “that the free exchange of ideas at the agency is diminishing, and that internal dissent is no longer being tolerated by senior managers – particularly when it comes to discussions about the 2011 census and new national household survey.” He goes further by saying: “a lot of good can be offset if you get one big thing wrong – and the big thing in this instance is census and NHS.”
At least we know the two people at Statistics Canada (well, formerly at StatsCan) with some sense of public service and responsibility. Now that they’ve left, StatsCan should be free from such annoyances.
Long form census data is worthless. The efforts to gather it, a waste of taxpayers money. Minister Tony Clement claims that he has his magic mojo working, enabling StatsCan to overcome the limitations of the mathematical and statistical sciences. I hope he shares it with the rest of us.
One question remains though, who will be rewriting all those science textbooks?