Canadian Politics: Munir Sheikh’s Essay “Good Data and Intelligent Government”

In the comments on last week’s post, some of you said you wanted to read more about Canadian politics on Persephone, which is great! So I’ll be writing regularly about what’s going on in politics here, and between omnibus crime bill, the gun registry, the NDP leadership race, and the upcoming Ontario election, there’s certainly a lot of potential material. But this week, let’s step back a bit and talk about the essay written recently by the Former Chief Statistician Munir Sheikh (pdf) criticizing the government’s decision last year to scrap the long form census.

(Editor’s note: I promise to make a more appropriate header for our Canadian political posts very soon. Hold your thumb over Lady Liberty until I do. ~Selena)

Non-Canadian Recap: Statistics Canada is (was, really) one of the most acclaimed statistical agencies in the world, with an international reputation for excellent data and efficiency. Since the 1970s, Stats Can has used a mandatory short form and long form census: in each census, two-thirds of households received the short form census, with around a dozen basic questions, and one-third of household received the long form census, with more in-depth questions used to compile a picture of the Canadian population. The census was mandatory, with potential jail time for those who refuse to answer it, though no one has even been sent to prison over failing to complete the census. The Conservative government decided, under pretenses of complaints about privacy concerns and the lack of necessity of the state to know how many bathrooms Canadian households have, to scrap the long form census and replace it with a voluntary National Household Survey. The Chief Statistician Munir Sheikh resigned over the decision, saying that it would irreparably damage the quality of data collected, and his staying on would implicitly support that decision.

Last week, an essay by Mr. Sheikh was published by the Centre for the Study of Living Standards, as a part of a collection of essays titled “New Directions for Good Government in Canada.” In it, he meticulously lays out the objectives and priorities of Statistics Canada, explaining clearly how the agency goes about its business and makes data available to governmental departments, organizations and companies, and Canadians. He’s unequivocal about how, exactly, scrapping the long form census will affect Statistics Canada’s work:

First, it will affect the long form survey data. It is a statistical fact that a voluntary survey cannot become a substitute for a mandatory census because of uneven response rates from different population groups and different size geographic areas. Increasing the sample size cannot offset this problem. Hence, many data users including the federal gov- ernment will lose the data quality they need.
Second, to the extent that the long form census data provide a benchmark for other Statistics Canada surveys, the quality of data from these other surveys would deteriorate.
Third, there is now risk about the quality of the short form data and other data produced by Statistics Canada.

He’s clear also that the reason that the reason the government could scrap the mandatory long form census was because by legislation, the Minister has final say over the methodology used by the dpeartment. Traditionally, the department has operated independently and without Parliamentary interference, but this was only tradition, not law. Mr. Sheikh says firmly that this needs to be reevaluated, and the independence of Statistics Canada needs to be enshrined in law rather than tradition.

And it’s that governmental interference that drove me to write about this, even though it’s about a dispute from over a year ago. This is perhaps the clearest, most non-partisan illustration that this regime is foolhardy and intrusive. By eliminating a mandatory long form census, it is inarguably true that the quality and accuracy of the picture of Canada as a whole and as regions, communities, and cities, is deteriorated. Not only have they hampered provincial and municipal governments, community organizations, and research organizations in doing their work, they’ve hampered their own federal work as well. The decision to scrap the long form census was criticized across the political spectrum in Canada, and the justification about privacy complaints (when in reality, fewer than five complaints had been filed in as many years) rang very hollow. But it’s much easier to obfuscate and avoid responsibility when the picture of a situation is muddy or incomplete, so I think the political motivations behind it were obvious.

I honestly don’t have any great argument to put forward about this essay: it’s incredibly clear, straightforward, and easy to read, and doesn’t say anything that I find particularly controversial but doesn’t pull punches either (though they are expressed incredibly diplomatically). It is, however, a great insight into how Statistics Canada functions, so I wanted to pass it around to the Persephoneers.

Next week: the Ontario election campaign