The Long and the Short of It – Census and NHS Questions

The following will appear in the Hill Times’ October 18th edition.

The Harper Conservatives repeatedly banged Canadians over the head this summer with their minority viewpoint on Canada’s long-form census questionnaire.

The questions, they said, were intrusive, and the government coercive for expecting answers. Mandatory suddenly became an ugly word for the law-and-order brigade.

Interestingly, their actions on the census this fall speak louder than the words they stubbornly stuck by over the summer.

Canada’s short-form census is still mandatory, upon pain of penalty. What’s more, the Harper Conservatives have added two new questions to the short-form census.

And the replacement for the long-form census isn’t substantively different than the real McCoy.

They have kept virtually all the same questions from the 2006 long-form census questionnaire that they railed about over the summer. The only exception deals with questions about unpaid household work (which, apparently, is too sensitive a subject).

But the kicker is that the Harper Conservatives added even more questions to the replacement for the long-form census questionnaire – the National Household Survey. It is none other than the Harper Conservatives who chose to ask Canadians about their religion, commute times, paid child care, spousal support, and housing.

So much for the freedom fighters.

Tracey Lauriault, of datalibre.ca, has unveiled these surprising results by placing each question from the 2006 census questionnaires cheek-by-jowl beside the proposed 2011 approach by the Harper Conservatives.

(See the comparison of the new and old short-form questionnaire here and the comparison of the new and old long-form questionnaires here.]

The only meaningful change is that the questions are no longer mandatory to answer, which begs the only real question: why bother asking Canadians anything at all?

The National Household Survey puts Canadians on the hook for a $30 million hollow exercise that won’t tell anybody that uses census data anything usable.

As experts have already advised this government, a voluntary survey means some people won’t answer, and the non-response bias will pollute the findings for the rest.

That’s because, irony of ironies, the scale and location of the non-response bias cannot be assessed without the census.

So the information collected will be unusable at the local level, where census data has become instrumental over the past decade to target resources through place-based policies.

It will also be unusable at the macro level, because previous research – relying on census results – shows us which segments of the population are most likely to avoid answering questions, given a choice.

That would be poor people and marginalized groups, who tend to disregard voluntary surveys because of barriers such as language, disability, newcomer status and decades of poor relations between government and Aboriginal communities.

Research also shows that rich people are more likely to turn down requests for information. For example, an international review of wealth surveys found that the response rate to Canada’s 1999 Survey of Financial Security, which measures household wealth, was 76% overall, but much lower among those with high incomes (60%). Response rates have since dropped.

We can also guess that, given their positioning over the issue, most members of the Conservative caucus won’t likely be filling out this voluntary survey either.

The leader of the government in the Senate, Senator Marjory LeBreton, typifies the difficulty Conservative caucus members face in balancing their frequently-voiced hatred of answering questions with their desire for information.

During the first question period of this Fall’s Parliamentary session, Senator LeBreton said, with a certain degree of irony: “I am sure that Canadians, being the generous and interested people they are, will fill out the household survey quite willingly and probably more willingly now that they are not demanded to do so.”

In the next breath she said: “The issue is that thousands of Canadians, like me, are being harassed by some part-time person hired to harass people into filling out the long-form census. I am stubborn sometimes, and I decided I was not going to do it until I was so afraid that I would be thrown into jail that I finally filled it out.”

So which is it? Are Conservatives themselves willing or unwilling to answer these questions?

All summer long the Harper Conservatives expressed outrage that the government (that would be them) coerced Canadians to answer census questions on pain of jail time.

Never mind that no one has ever been ordered to go to jail for failing to fill out a census questionnaire. Back in the thick of it, Minister Clement issued a summertime promise to introduce legislation that would remove jail terms from the Statistics Act in the fall when Parliament resumed.

Canadians are still waiting for the government to make good on that promise. It would fit in nicely with a Fall legislative agenda that is crammed full of Conservative “crime and punishment” reforms. But it has yet to be tabled.

Apparently putting more Canadians into jail for longer is a higher priority than keeping them out.

Just one more incoherence in this government’s “strategy”, which pretty much sums up the long and the short of it all.