Industry Committee on July 27th, 2010

[Robert Bouchard, Bloc, Chicoutimi—Le Fjord, QC]

Thank you. We can see that you are very knowledgeable about the data from Statistics Canada and that the decision entails many repercussions.

In your opinion, Mr. Simard, are there people or groups who will be particularly at a disadvantage because of the implementation of the decision to make the long questionnaire voluntary?

[Martin Simard]

Absolutely. My colleague explained it very well a few minutes ago. When you ask people to fill out the form voluntarily, even if it seems to be completely normal and positive, it actually distorts the data in the census sampling process. In other words, the people who will answer are more likely to display specific characteristics. For example, they have more time or more education, so that they value the census or understand its importance better. So the data we get are not representative of the whole population, but rather of people who show goodwill or who like to take part in surveys.

So, at first glance, it is a small change that is well intentioned and can seem like a good idea. In fact, it is a technical change that will drastically alter the credibility and value of the census. At the end of the day, it is a rather technical debate, but it is still quite important given the repercussions on the quality of census data.

[Robert Bouchard, Bloc, Chicoutimi—Le Fjord, QC]

So, if I understood your comments correctly, the government's decision would significantly compromise the quality of the data collected by Statistics Canada.

[Martin Simard]

Precisely. I will say it again. At first, we could say that, given the concerns expressed by some Canadians, we will not make all forms mandatory. That may seem easy. We do not abolish the forms completely, but we make the responses voluntary. That may seem like a rather minor change and like something that finally makes sense .

But actually, on a technical level, that would change the sampling process fundamentally. The people who will answer will not necessarily be representative of the whole Canadian population. So that will seriously compromise the quality of data. This leads us to say that everyone using the data, in both private companies and scientific research, will be affected by these changes.

[The Chair, Conservative, Michael Chong]

Thank you very much, Mr. Bouchard.

Thank you, Mr. Simard.

Mr. Bernier.

[Maxime Bernier, Conservative, Beauce, QC]

Thank you, Mr. Chair.

I would like to take this opportunity to answer one of my opposition colleagues' questions.

This morning, my Liberal and Bloc colleagues said there were no problems in Canada, that 95% of Canadians filled out the long, detailed questionnaire on their private lives. They said that this debate was useless, that everything was going well and that people were happy to fill out the 40-page detailed questionnaire. What I would like to say to the members of the opposition is that people fill it out because they are threatened, either with a prison sentence or a fine. If the opposition members want to improve the participation rate, I suggest that they increase the threat, so that Canadians have to pay a $5,000 fine rather than $500, and that the prison sentence be three years rather than three months.

Mr. Chair, that is not the issue. It is really simple. It is a question of principle. In my opinion, people should be free to answer or not answer questions about their private lives. I would like to hear Mr. Doucet's opinion on that.

In a democratic society, do you believe that it is normal to threaten people with a prison sentence and a $500 fine if they refuse to answer questions on, for example, how many bedrooms they have in their houses or the time they leave their house to go to work?

[Bradley Doucet, English Editor, Québécois Libre]

It's normal in the sense that it's widespread, but I don't think it's normal in the sense that it makes sense. I don't think it should be that way.

I agree with the government's position that if people need to be forced to provide this information, then there's something wrong there. I don't think it's right that if someone refuses to fill out the long form, he or she has either a fine or a jail term. If the people who are supposedly being helped by the information collected from the census can't be persuaded to fill it out, I don't think they should be forced to under penalty of law.

[Maxime Bernier, Conservative, Beauce, QC]

Thank you very much.

Monsieur David Tanny, what is your position on that?

[David Tanny, Associate Professor, Department of Mathematics and Statistics, York University]

I'd like to speak from two points of view.

First of all, I don't think this is a technical debate. The reason I don't think it's a technical debate.... There is the issue of the statistics. I'm not going to deny, as a statistician, that if one makes the form voluntary without any other incentives, then in fact there will certainly be a change in the response rate. One should know, however, that there has already been a change in the response rate. One should also know that there are other factors that have to be considered.

In 2006, Statistics Canada went to using the Internet. There are many people, for example, who don't know how to use the Internet and simply turn it over to their children. So we don't know exactly the quality of the information we're getting once the information is turned over on the Internet. It could be a 15-year-old going ahead and putting in the information rather than an adult, as we were doing in the past.

Now, I agree with the idea or the general principle that using the Internet is a less costly way and probably a much more efficient way of doing it, but what I'm saying is that there are other factors that come into play with regard to the quality of the information and how much people can rely on the information we have.

A second issue is that Canada is one of the few nations—a handful of nations—that actually do a full census every five years. There are rolling censuses being done continuously. The Americans have moved to what's called the “American community survey”. We have a national household survey.

The American community survey, of course, is going ahead and every month sending out a certain number of forms and requiring it. It is mandatory in the United States. I disagree with that, though. There is such a thing known as negative bias that does occur when you make a survey mandatory. There are studies on negative bias, and the question is, how do you overcome it?

Well, let's just take a look. Here is our survey. On one end of the scale you have the penalties. If you want to make sure that everybody does a particular survey, make it 10 years in jail, 10 years for not answering any questions--

[The Chair, Conservative, Michael Chong]

Professor Tanny, I think Mr. Bernier wants to ask you another question.

[Maxime Bernier, Conservative, Beauce, QC]

I think you're absolutely right. As I said at the beginning...just increasing the penalties and people will answer all that.

But I have a question for Monsieur Simard.

From what you said earlier, you are in full support of

the long survey, mandatory census,

for everyone who will receive the questionnaire.

What questions should be mandatory? Should people have to answer the question on what time they leave their house to go to work, under threat of being fined or sent to jail?

[Martin Simard]

We could examine the questions and filter them out.

[Maxime Bernier, Conservative, Beauce, QC]

Do you think we should send people to jail if they do not answer this specific question?

[Martin Simard]

You are indeed right. We should not do that.

[Maxime Bernier, Conservative, Beauce, QC]

Should we send them to jail if they do not reveal how many bedrooms they have?

[Martin Simard]

No, but...

[Maxime Bernier, Conservative, Beauce, QC]

Thank you. You have answered my question. There are more than 49 questions, and I would like to know your opinion. We think that people are free and independent and that they should not be threatened with a prison sentence.

Should we send people to jail if they do not tell us how many bathrooms they have?

[Martin Simard]

If you'll allow me to answer, I'd say that the prison sentence could be re-examined. I agree with you on that point, but as for the citizens' duty to fill out a census form, I don't really think that it's asking too much. In some provinces or countries, they start each morning with a prayer...

[Maxime Bernier, Conservative, Beauce, QC]

According to you, in order to reduce the margin of error, statisticians prefer that people who receive the questionnaire be obligated to complete it, under penalty of fine or imprisonment, rather than respond to an optional survey, which would respect their freedom.

You said earlier that the first option improves the quality of the data. So, to get the best possible data, the Cadillac of data, you would like Canadians to be threatened with imprisonment. Is that right?

[The Chair, Conservative, Michael Chong]

Thank you, Mr. Bernier.

[Martin Simard]

I said that imprisonment struck me as an extreme measure, but perhaps that is a separate issue.

That being said, I don't think that...

[Maxime Bernier, Conservative, Beauce, QC]

But that is precisely the issue today, Mr. Chair. Are Canadians and Quebeckers free to answer the questions with this threat hanging over them, or not...

[The Chair, Conservative, Michael Chong]

Thank you, Mr. Bernier.

[Martin Simard]

Allow me to finish my answer.

We should keep in mind that answering census questions is not a task that is very difficult or unreasonable. It should be seen as a citizens' duty rather than as a possible source of punishment.

[The Chair, Conservative, Michael Chong]

Thank you.

Mr. Gravelle.