Tagged: Statistics Canada
Statscan v. US Census Bureau
Contrarian reader Wallace McLean noticed something else about those maps:
[T]he US Census Bureau seems to generate unemployment data for the 3,140 counties and “county-equivalent” units of geography below state level, with an average population of under 100,000.
Statistics Canada only provides (roughly) comparable data for 73 “economic regions” within Canada, with no sub-provincial/territorial data for PEI or the territories. The 73 regions have an average population of over 450,000.
Even if you could get free and up-to-date data out of Statscan, it’s not nearly as fine-grained as what they seem to have in the States. There would seem to be some fundamental methodological difference in how the two stats agencies approach measuring employment and unemployment.
Census: a mistake in the long form
Contrarian reader Gus Reed has found a mistake in Census Canada’s long form questionnaire — or at least in the sample that appears on the agency’s website.
It seems to me that at the top of page 5 the columns should be labeled “Person 3,” “Person 4,” and “Person 5″ – continuing the logic of page 4. Is it my PDF reader that’s wrong, or did StatsCan send out 2.4 million errors?
StatsCan hasn’t sent out anything yet, and there’s still time to fix the error, along with the much more serious mistake of making the long form voluntary. But Gus is right. The headers are inconsistent in a way that might lead respondents to omit information about some household members, or else enter responses in the wrong place. It’s hard to describe the problem without looking at the form, but this ought to be fixed.
Gus has a lot more to say about the long form. Stay tuned.
HARPER DEFEATS CENSUS
Harper spokespeople argue that sending the voluntary census long form to a larger number of people will compensate for any loss of data quality due to the newly voluntary nature of the form. Milan Ilnyckyj explains the fallacy.
One of the biggest challenges in statistics is collecting a representative sample: finding a subset of the population that will do a good job of approximating the whole group. When a dataset contains a lot of sampling bias and is not reflective of the general population, it is essentially worthless as a guide. That cannot be fixed by using a larger sample size, nor can it be dealt with via fancy mathematics.
The classic example of sampling bias is the ‘Dewey Defeats Truman’ headline, from the Chicago Tribune in 1948. The newspaper got their prediction wrong because they sampled people with telephones, at a time when telephones were comparatively rare. Most of the people who had them were rich, and rich people were more supportive of Dewey. As a consequence, telephone polling provided bad information about the likely voting behaviour of the whole population.
While on the census fiasco, Jim Brown, guest host of CBC Radio’s The Current, was uncharacteristically ill-prepared this morning for his interview with Conservative sock-puppet Tim Powers. He let Powers float unchallenged from one specious talking point to another, even letting him equate the supposed intrusiveness of a standard census question about the number of bedrooms in a respondent’s house to Pierre Trudeau’s decision to repeal laws outlawing private homosexual acts. If you’re going to guest host a national show, you need a passing familiarity with recent Canadian history, and you need to bone up on the issues of the day. Brown is usually better than this.
Letter of resignation
The libertarian devotion to individual freedom that led the Harper Government to kill Statistic Canada’s mandatory long form census questionnaire apparently did not extend to the Chief Statistician of Canada’s letter of resignation.
Munir A. Sheikh posted a note about his resignation on the agency’s website late Wednesday night. The Harper Libertarians redacted it Thursday morning, replacing it with an uninformative generic message.
Here, for the record, thanks to Kady O’Malley, is the full text of the Chief Statistician’s censored message to Canadians:
July 21, 2010
OTTAWA — There has been considerable discussion in the media regarding the 2011 Census of Population. There has also been commentary on the advice that Statistics Canada and I gave the government on this subject.
I cannot reveal and comment on this advice because this information is protected under the law. However, the government can make this information public if it so wishes.
I have always honoured my oath and responsibilities as a public servant as well as those specific to the Statistics Act.
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. I want to thank him for giving me the opportunity of serving him as the Chief Statistician of Canada, heading an agency that is a symbol of pride for our country.
To you, the men and women of Statistics Canada – thank you for giving me your full support and your dedication in serving Canadians. Without your contribution, day in and day out, in producing data of the highest quality, Canada would not have this institution that is our pride.
I also want to thank Canadians. We do remember, every single day, that it is because of you providing us with your information, we can function as a statistical agency. I am attaching an earlier message that I sent to Canadians in this regard. In closing, I wish the best to my successor. I promise not to comment on how he/she should do the job. I do sincerely hope that my successor’s professionalism will help run this great organization while defending its reputation.
Munir A. Sheikh
