The census issue won't go away. Ever.

The government modified its census position yesterday. Unfortunately for the cause of responsible government, these changes don't alter the situation much - but they do illustrate the vapidity of the government's stance.

Firstly, in the face of a court challenge by minority language groups, the questions on language use are now going to be part of the mandatory short form. The reason for this is clear: there was no way that the government could convince the courts that a voluntary survey would provide information that would be credible enough to fulfill its obligations to minority language groups to provide services in the official language of their choice.

This is of course an implicit recognition of something that critics have been saying explicitly for the past six weeks: that the voluntary survey will not produce reliable data. What's even more astounding is that on very the same day, Tony Clement saw fit to repeat the claim that increasing the sample size would fix the self-selection problems associated with the voluntary survey. If the government really believed that assertion, there would be no need to move the language questions to the mandatory short form.

The other thing the government did was to announce legislation to remove the threat of imprisonment for those who do not respond to mandatory surveys such as the census and the Labour Force Survey. The point has been made before that if the government did not want to threaten people with jail for not answering certain questions, the remedy was to remove the questions and/or remove the threat of jail; making the entire long form voluntary was a non sequitur. Now that the threat of imprisonment has been removed, then there really isn't anything left of the government's original case for making the long form voluntary.

Yet the decision still stands, and we need to look forward. It's an ugly, ugly picture.

Before the decision went public, Statistics Canada estimated that the response rates for the long form would fall from about 95% to 50%, but it could be brought back up to 70% if they spent a lot of money pestering non-respondents. This result was of course deemed to be unacceptably low a few months ago, but even this level of failure seems like an unattainable dream now. Now that the census has been made a playground for partisan politics, many CPC partisans will decide that their party's cause will be best served by not filling out the voluntary forms, and many opposition supporters will boycott the long form in protest. The long form data will be a dog's breakfast.

And there will no doubt be many people who will have missed the distinction between the short and long forms and will wrongly assume that they can pitch their short forms into the recycling bin without consequence. After making such a big deal about not forcing people to fill out the census form, the government will be hard-pressed to justify using the powers of the State to force citizens to fill out the short form. There's a real risk that the short form data won't be usable, either.

These are problems that will explode in the government's face over the next few months. But it won't end there. As the months and years wear on, every single census release will be accompanied with a lengthy discussion of to what extent changes since 2006 reflect reality or the failed 2011 census. This is going to go on for years; a cursory search kicks up at least 11 discussion papers based on the 2006 census published by Statistics Canada in the past 12 months. (See here for a summary of output from the 2006 census.) And that's just Statistics Canada. It doesn't include studies done at other federal departments such as Finance or HRDC, or by other agencies at other levels of government.

Of course, non-government researchers will have to devote any number of person-years dealing with the wonky 2011 numbers. Many future studies will no doubt be obliged to use a binary indicator variable to capture the 2011 outlier; this indicator will be known as the Clement Dummy. There will be snarky variations on this notation during seminars.

The implications will go beyond empirical studies that make direct use of quinquennial census data. Variations in higher-frequency numbers will also be suspect:

• The Labour Force Survey uses the census to make sure that its rotating panel is representative of the working-age population. As time goes on, our measures of what that means will deteriorate, and the quality of the LFS data will deteriorate as well. This matters; a small thing such as wonky seasonal adjustment in the education sector was enough to generate stories about whether or not the Bank of Canada should pay attention to the recent employment numbers when deciding the short-term course of interest rates.
• Statistics Canada uses weights derived from the Survey of Household Spending to calculate the Consumer Price Index, and the SHS uses census data to make sure its sample is representative of the entire population. If those weights do not accurately reflect Canadians' spending patterns, then variations in the CPI will not accurately reflect variations in the cost of living.

These are only the most visible manifestations, and readers are encouraged to add their own personal favorites to this list.

What is even more discouraging is that even if the 2016 census is run properly, the problems associated with the 2011 census can't ever be fixed. Time series data are constructed by piecing together different survey results, and a chain is only as strong as its weakest link.

In many important ways, our history is about to become a lot shorter.

Comments

[Ed Seedhouse]

Well if, as you suggested, many Tory supporters refuse to fill in the long form to show their support for the government, then their particular group will be badly undersampled and any number of programs that rely on this information will them serve them less and concentrate more on the groups that completed the form. So they seem to be digging their own graves here.

[Appalled but not confused]

For what it's worth -- viz., $25-M -- it seems the Con. braintrust belatedly figured out that all the trashing of the Census & StatCan they've precipitated has greatly endangered the short form, too, whose comprehensive mandatory head count definitely _is_ req'd by law & which directly affect both the Billions in transfer payments & the electoral boundaries they want to rejig.

So acc. to Ottawa's trade paper the Hill Times** (the pols' v. of 'Variety Magazine'), they're now reallocating that extra $25-M ($5-M is still needed to print & post all the materials needed to increase the NHS sample by 50%) in additional advertising for.... the need for and wonderful sense of civic pride we'll get from filling out the _short_ form. And there'll be no increase in the staff or budget to try to do any follow-up on the NHS. So don't expect its response rate to be any better than, say, 40%.

So, if the interested parties can't get the courts to make it mandatory to meet the remaining 34 pieces of federal legislation that are indexed to Census data,* they should just cancel the NHS altogether this go-round until the dust settles, and save the over $100-M I've heard was budgeted just for it.

*which are listed in on p. 10 of the 11 pp. Statcan Backgrounder that Kady O'Malley's posted at the end of:
www.cbc.ca/politics/insidepolitics/2010/08/census-document-dump-things-fall-apart-the-centre-cannot-hold.html )

** www.thehilltimes.ca/page/view/integrity-08-02-2010

[Insert Real Name]

To this non-academic/non-specialist, whether or not they effectively dig their own graves is beside the point, which is that one of the basic sources of decision-making cross-country data in Canada is compromised. And for what end? To satisfy bats-in-the-belfry ideological prejudices from a government that is always trying pose as responsible government! In our country we actually do need good census data, perhaps more than some countries who, through history and accident, are more coherent internally.

Perhaps Canada really will become some sort of 21st century analogue to the political dissolution of Austria-Hungary: riven by centrifugal regional antipathies and political myth-making, fractured by steadily increasing but ill-understood social disparities and divergences, etc., etc.

[Appalled but not confused]

p.s., that table of fed. Acts indexed to census data has now been reproduced at the middle of this page:

http://www.progressive-economics.ca/2010/08/12/the-medium-form-is-the-message/

[Wilf]

There has been so much misinformation on this issue it is not even funny. The comment by Ed Seedhouse

"So acc. to Ottawa's trade paper the Hill Times** (the pols' v. of 'Variety Magazine'), they're now reallocating that extra $25-M ($5-M is still needed to print & post all the materials needed to increase the NHS sample by 50%) in additional advertising for.... the need for and wonderful sense of civic pride we'll get from filling out the _short_ form. And there'll be no increase in the staff or budget to try to do any follow-up on the NHS. So don't expect its response rate to be any better than, say, 40%".

The misinformation is that there NEVER was any additional funds allocated to get cooperation of Canadians on the voluntary NHS, as Minister Clement stated many times in the media. From the start, the $25 million was to ensure the integrity of the short form results that are used to distribute billions of dollars to provinces, cities, municipalities, etc.

Check the CBC posting of released emails on the Census. There it is stated by Stat Can official that there are no funds allocated for NHS advertising or extra follow-up efforts by Stat Can. This fact has been also confirmed with knowledgeable insiders.

As it stands now, the NHS should be cancelled. This will save some $60 million plus some of the extra $25 million for advertising, etc. The initial 2011 Census budget included a significant reduction in paid advertising from the 2006 level. So some of the $25 million is just a replacement bacl to 2006 levels.

More importantly, the say 2.5+ Canadian who do fill out the NHS will not be wasting there time providing information, that because its non-representativeness of the whole population, will be not be usable. This will create respondent anger that will negatively impact our statistical system for years to come.

No data,lousy data, misleading data, anger in the respondent community that can lead to lower overall response rates, loss of credibility of Stats Can information. Are these the real reasons for the governments approach to the 2011 Census? Prob(Yes)= 0.90 at 99% confidence level is my estimate. As confirmation, consider the following:

A very interesting article published in one of the national papers was titled, if I recall correctly, " Getting Stats Can off the salt". It quotes then Minister responsible for Stat Can, Maxime Bernier, saying that if you cut back on Stat Can then departments can not use their info to develop new programs and spending more dollars. Cutting Stat Can leads to smaller government. Also reported was, when Bernier was Foreign Affairs Minister, he proposed big cuts to Stat Can. This was supported by the PM but not by Kevin Lynch the PCO Clerk.

With the Census cut, dropped surveys, the cancellation of the PALS post-censal survey, cuts in the analytical activities and other developments, the Harper-Bernier plan has been implemented over the last two years. This happened under the watch of the previous Chief Statistician. The statistical future in this country is very uncertain.

[Karen Krisfalusi]

Well I'm still optimistic the media is going the win the day on this one. They aren't letting up and it's very encouraging to see! Reading The Globe and Mail is a joy! Jane Taber headlined a report on Flaherty's policy retreat and called "Jim Flaherty's Census Defense". The editorial powers that be are warmed up and journalists are full of contempt, inserting census references everywhere. John Ibbitson wrote today Long or short, Tories must retreat and warned that if they don't "this thing will still be making headlines when the leaves begin to fall." So chin up!

[Karen Krisfalusi]

Oh I made a mistake. Ibbitson wrote in July. Sorry -- not sure why that article popped up in my Google Reader today. Still, today's editorial in the Globe was critical. It's a very hard policy to battle and it will just take a little more time to get it done. I think that Harper's mindset is key to winning the spin. Here's a guy is buying fighter jets, talking about continuing the Afghan engagement and expanding prisons. There is a corner of that mind that doesn't want to be seen to be abusing people with government powers because he is now so very powerful. He has to be shamed into recognition that his electoral politics ambitions with this are going to play front and centre. He must be made to understand that what he used to justify this policy change for electoral gain (his reluctance to be percieved as abusing government power) was a personal moral crime. He's coming undone from the inside.