Stats Canada to make long-form census voluntary

[Tapioca]

This announcement was made last week and I must admit, I glanced over the headline and didn't think much of it. Here's an article from last week on it:

Tories scrap mandatory long-form census

StatsCan says quality of data will suffer

Jennifer Ditchburn

Ottawa — The Canadian Press
Published on Tuesday, Jun. 29, 2010 6:39PM EDT

Last updated on Wednesday, Jun. 30, 2010 7:08AM EDT

The federal government is scrapping the mandatory long census form in favour of a voluntary survey – a move some critics blame on a Conservative campaign to slash analytical work done by Statistics Canada.

For the first time in 35 years, the census will not feature a detailed, long form that Canadians are obliged to send back to the government.

Instead, a mandatory short form will go out to everyone for next year's census, with basic questions about how many people live in the household and their ages and genders.

The voluntary “national household survey,” with detailed questions about ethnicity, income and education, will be sent to one-third of homes. That's an increase from the 20 per cent of homes that used to get the mandatory long-form.

The move is a response to protests from some Canadians who resented the personal questions in the long form. Similar opposition has been raised in the United States by some Republicans opposed to Washington collecting and analyzing data.

“Our feeling was that the change was to make a reasonable limit on what most Canadians felt was an intrusion into their personal privacy in terms of answering the longer form,” Erik Waddell, spokesman for Industry Minister Tony Clement, said Tuesday.

Some Canadians also criticized Statistics Canada for purchasing data software from U.S. defence corporation Lockheed Martin, although the agency has insisted the firm does not have access to personal data.

One woman in Saskatchewan is still battling the federal government in the courts over her refusal to complete the census.

Don Rogers, a Kingston, Ont., man who mounted the “Count Me Out” campaign against the census, counselled Canadians on how to provide minimum co-operation on the census. That included crumpling the forms or writing answers upside-down.

“The Canadian census was originally a count of heads and cows ... and we would certainly like it to revert back, if not all the way to that point, but to a collection of very basic, minimal information,” Mr. Rogers said.

“There's no need to ask how many rooms in your house, there's no need to ask how many hours of unpaid work do the members of your household do.”

What the impact will be on the quantity of data that Statistics Canada receives is a question even the agency can't answer. Officials are counting on Canadians, who generally have high rates of response to surveys, to do their civic duty and mail in the forms.

“What we can tell you is that the data we will release will be of quality, but we do acknowledge that we may not get the same level of detail as that of a census,” said Rosemary Bender, director-general of the agency's census, social and demographic statistics branch.

“This is something we'll be monitoring closely.”

Insiders who spoke to The Canadian Press on condition of anonymity decry a new world order within the agency since the Conservatives came to power in 2006 and legendary chief statistician Ivan Fellegi retired.

Employees were told a little over a year ago that there would be less emphasis on analysis. A highly praised survey on immigrants to Canada, for example, has been axed. Other analytical jobs, in areas such as business and trade statistics, and the aging population, have been eliminated.

Some employees say the agency will lose its status as the best statistical office in the world.

One Statistics Canada source said the move could have a negative impact on the dozens of provincial governments, community groups and other organizations that depend on the data for developing policy.

“It will be a disaster. A lot of policy across Canada has been based on that long form,” the source said.

University of Western Ontario statistician David Bellhouse says the main problem with making the long form voluntary is that there will be a bias in the numbers.

Certain groups of Canadians might be more inclined not to fill out the form, skewing the results. Already, many aboriginals living on reserves have balked at filling out the census.

“You're getting an incomplete picture. The people who refuse to respond might be different from the people who are willing to respond,” Prof. Bellhouse said.

“In surveys, it's generally known that when you ask the question about what is your income, people in higher incomes typically refuse to respond to that question and you can be slightly biased downwards.”

[gdoh]

our decision makers dont give a fuck about the average working class

[q0192837465]

Nice, i wonder if we'll still get a representative sample now. Doubt it tho.

[RRxtar]

well. its not nearly as bad as this article makes it sound i think?

they are sending the census out to 33% of homes, up from 20% of homes from the previous sensus. so if like 3 out of every 5 people who receive the voluntary census fill it out, it will still end up with the same size sample group.

[q0192837465]

^ it's not about the sample size, it's about the representative sample. The gov't will only get data from people who are willing to fill out the survey. THIS is the problem.

[chouchou]

^and usually those who fill out the data have some sort of opinion in regards to certain questions. bias bias bias

[TRDood]

bad move.

a lot of research is based out of stats can census data, if they decide to throw detailed variables out the window, then there will be a big job loss in the research and policy areas...

[subordinate]

Originally Posted by TRDood

bad move.

a lot of research is based out of stats can census data, if they decide to throw detailed variables out the window, then there will be a big job loss in the research and policy areas...

yep, very bad move. I'm sure everyone hated filling out the damn thing; it was like a damn book.

Although time consuming, I'm sure the statistics were used for a beneficial purpose.