Industry Committee on March 8th, 2011

[Mike Lake, Conservative, Edmonton—Mill Woods—Beaumont, AB]

Well, I'll leave my point of order now. I have questions I want to ask.

The clause that I'm referring to talks about the questionnaire conforming substantially in length and substantive scope to the long-form census used to take the census in 1971. In terms of differentiating between the 2006 census and the 1971 census, I would assume there was a specific reason that 1971 was chosen. I think there are some pitfalls when we talk about having substantively the same long-form census that we used in 1971. For example, in question 2, “Relationship to Head of Household”, there's a statement in the 1971 census that says the head of the household is the husband rather than the wife, the parent where there is one parent only with unmarried children, or any member of a group sharing a dwelling equally. In the substantive question in question 2, the head of the household has to be the husband rather than the wife. I think there are some pitfalls choosing the 1971 census as the one that we would use substantively.

[Carolyn Bennett, Liberal, St. Paul's, ON]

Mike, I would like Mr. Fellegi's advice as to whether there should be an amendment to change it from 1971 to 2006 in the bill.

[Mike Lake, Conservative, Edmonton—Mill Woods—Beaumont, AB]

In fairness, I'm asking you as the mover of the bill to explain your bill.

I'll move on to--

[Carolyn Bennett, Liberal, St. Paul's, ON]

No, I'm actually asking you as a parliamentarian if you'd like to make an amendment to that.

[Mike Lake, Conservative, Edmonton—Mill Woods—Beaumont, AB]

Well, we'll go clause by clause, I'm sure, at some time in the future here.

In your first clause of the bill, you talk about the distribution of the questionnaire...to at least 20% of all households or whatever percentage of households is determined to be necessary by the Chief Statistician to ensure an accurate statistical representation...

There's no mention of cost there. You've left it absolutely wide open. It's any percentage between 0% and 100%, at the discretion of the chief statistician, with absolutely no mention of cost. Is there a reason you didn't mention cost? Have you done a costing of the potential implications of that clause?

[Carolyn Bennett, Liberal, St. Paul's, ON]

We know right now that the minister has announced that in order to do the greater survey, he is spending $30 million more than what the mandatory survey would have cost, and we have already heard that doing the larger sample size does not fix the bias problem.

[Mike Lake, Conservative, Edmonton—Mill Woods—Beaumont, AB]

So what you've just answered--

[Carolyn Bennett, Liberal, St. Paul's, ON]

What we're saying is that this bill will save the Government of Canada $30 million.

[Mike Lake, Conservative, Edmonton—Mill Woods—Beaumont, AB]

What you've just said is that the minister actually did some research on costs, but I don't believe that you have, unless you could table the research that you've done on the potential costs of your bill from 0% to 100%.

[Carolyn Bennett, Liberal, St. Paul's, ON]

What we're saying is that the minister is saying he's going to spend $30 million more in order to get the larger sample size on the voluntary survey.

[Mike Lake, Conservative, Edmonton—Mill Woods—Beaumont, AB]

Maybe I'm not being clear. I'm asking about your private members' bill and the research that you might have done in preparing the private members' bill.

We'll leave the tabling of those costs. You could table them at a future date, with the range from 0% to 100% of all households, because that's the range that applies, according to your bill here.

[Carolyn Bennett, Liberal, St. Paul's, ON]

You know what, Mike? This is about information that is the navigation system for our country, and we know that doing a larger voluntary survey is more costly and less accurate than doing a smaller mandatory survey.

[Mike Lake, Conservative, Edmonton—Mill Woods—Beaumont, AB]

I have one last question. The final clause of your bill, clause 2, says that “...for every refusal or neglect, or false answer or deception” there's a fine not exceeding $500. To clarify, someone—a new Canadian who doesn't want to tell the government what their religion is—would be subject to a fine up to $500 because they don't want to tell the government what their religion is. What happens if they don't pay the fine? What does the law say, just to be clear, on what happens if someone decides that they don't want to pay the fine because fundamentally they don't believe they should have to tell the government what their religion is?

[Carolyn Bennett, Liberal, St. Paul's, ON]

I think, Mike, it would probably be the same as it is in the present law--imprisonment--and that has never happened.

We have laws, and there are penalties, and we do our best within that. It has been up to the cabinet always to determine what the penalties and what the questions are. Getting rid of the mandatory long-form census is our problem.

[The Chair, Conservative, David Sweet]

That's the time, folks. Thank you.

We have about two minutes left.

Monsieur Bouchard, Monsieur Cardin, we have time for one more question.

No? Okay.

Then we have time for one more question from the Liberal Party.

[Dan McTeague, Liberal, Pickering—Scarborough East, ON]

Mr. McKinnon, there is a question that I think has all of us here somewhat perplexed. Your organization in several months is going to have to pick up the pieces and make an analysis of what has happened. I say “pick up the pieces” in terms of an assessment of how effective this is going to be.

Can you anticipate for us what effect this change will have in terms of real outcome and in terms of the reliability of the data? It's $30 million later, and we have nothing to show for it.

[Ian McKinnon]

Mr. McTeague, you may be under some misapprehension about the National Statistics Council. We're an advisory body to the chief statistician. He may pose questions and ask for our advice on certain things, but the issues you're talking about are highly technical and detailed. They involve a great deal of work that will be done within StatsCan.

[Dan McTeague, Liberal, Pickering—Scarborough East, ON]

I appreciate that, Mr. McKinnon, but you are going to be in a position of giving advice based on what you perceive. You are experts in this field. The reliability of the information is going to be critical. The information you're going to give will be based on assumptions as to whether the data are going to be accurate or not, based on whether there is a mandatory form or not.

I'm simply asking if your group, as experts, will be in a position to assess, and how soon will that assessment take place?

[Ian McKinnon]

The assessment will be conducted by Statistics Canada. They may ask for our advice on some element of it, and the typical post-census work would begin as data collection ends, I assume.

I would actually have to turn that question over to the two gentlemen from Statistics Canada, who would know a bit better.

[Wayne Smith]

Statistics Canada will obviously validate all of the data from the 2011 census to the extent that we can, and we will publish the results of what we find. If we find data quality problems, we'll make them known. If we find a variable that in our view is so seriously defective as to be misleading, we will not proactively publish it. The data will remain available to the people who want to work with it, but we wouldn't proactively publish it. We don't expect that, but it could potentially happen. It has happened in the past.

In any case, as we go about the publishing, we will be carrying out quality assessments on the data and we will make that information available to the Canadian public.

[Dan McTeague, Liberal, Pickering—Scarborough East, ON]

If the utility of this turns out to be significantly lower than would be tolerable by any comparison to the past or to any other jurisdiction that has a similar mandatory aspect, I'm wondering whether you'll have a timeframe in which you'll be able to make an assessment, and what that timeframe will be, Mr. Smith.

[Wayne Smith]

The quality assessment program will be carried out over the next few years as we go forward. We are to some extent reworking it because of the changes to the census and the introduction of a voluntary national household survey. I don't have the specific dates here for when we'll publish, but we will publish—

[Dan McTeague, Liberal, Pickering—Scarborough East, ON]

Thank you, Mr. Smith. I appreciate that this was a political decision imposed on what would otherwise have been within StatsCan. There is no sense in changing what has always worked; if it wasn't broken, why would you want to fix it?

I take it that you'll then have a period of time in which you're going to make an assessment, down the road, as to whether it has in fact worked, and that iIt may be several years. Is that what you're saying?

[Wayne Smith]

On every census we conduct a quality assessment. We will be carrying out the same kind of quality assessment for the 2011 census and the national household survey, making the results available as quickly as we can and making the data available, again as quickly as we can.

[Dan McTeague, Liberal, Pickering—Scarborough East, ON]

The Conservatives are saved by the bell.

Thank you, Mr. Smith.

Thanks, Chair.