I wrote in an earlier blog post that the rest of the data coming from Stats Canada would be called into question if the hyperventilating from the experts cited by other media were true.
From the earlier post
Postmedia, the people that used to be called CanWest, quote Ivan Fellegi, a man who worked at Stats Can for 50 years, 23 of them as the Chief Statistician. “Every ounce of my professional knowledge, which is certainly the same professional knowledge that is very widespread at Statistics Canada, tells me that this is statistically absolutely wrong.”
If that is true then most of the data that Stats Can produces is useless. Every month the unemployment rate and GDP numbers are issued and governments and markets react. Apparently that is foolish, those numbers are not based on mandatory forms sent to 20% of all Canadians, and going by what the experts say, that means it is all skewed data.
Now, thanks to a fellow on Twitter known as @lukasneville I stand corrected.
Am I required to participate in the survey?
Yes. Parliamentary legislation has made it mandatory for people to participate in this survey because of the vital information it provides about Canada’s labour market.
So the Labour Force Survey is mandatory. So this does change my argument somewhat but not substantially.
The employment numbers are required. The Canadian Health Measures Survey is voluntary. The New Vehicle Motor Sales survey in mandatory. The Canadian Community Health Survey is voluntary.
So it remains that some data from Stats Canada is more reliable than others. Some is voluntary, some is coerced. Note to CMAJ, the health info is voluntary.
A number of you have written, some to call me names, some to say that the long-form census is the data used to guide sample size or determine sample weights in voluntary surveys. Fine, I do understand that despite the regular questioning of my intelligence. Still, the key questions for sampling are found in the short-form census that we all fill out. As for incomes, a far better source for that data, in aggregate, is from our tax filings.
It really is odd. Just recently in Ottawa many members of the national media were cheering when the privacy commissioner launched an investigation into Facebook, a private company none of us are forced to deal with. Now with the government promising greater privacy in dealing with government, something none of us are allowed to avoid, this same national media act as offence against the government, calling for greater intrusion.
Oh, and as for the Labour Force Survey being mandatory, Tony, that’s something else for you to fix.
Tags: 2011 long-form, census
Taking a view that opposes the Liberal worldview? Wait for it, “you must be in the pocket of big oil” and a redneck.
Free loaders, long time employees that have a vested interest are complaining, who would have imagined it?
Thanks, Brian, for making the correction. There are merits to both sides of this debate, but it’s important for people’s positions to be shaped by accurate information.
(It’s also relevant to note that the LFS is not only mandatory itself, but uses long-form census data in a number of ways, including the use of census income data to establish sample stratification. The reliability of unemployment data hinges on both the mandatory LFS and the mandatory long-form census.)
Thank you again for updating the site with the correction.
All the best –
Lukas
Are you really so stupid that you can’t see the difference between a private company that has demonstrated it would only respect people’s privacy after a huge outcry and the threat of legislation, and a government survey which has a long history of privacy?