Industry Committee on July 27th, 2010

[Anthony Rota, Liberal, Nipissing—Timiskaming, ON]

For the short-form census and the agricultural census, what is the penalty?

[Tony Clement, Conservative, Parry Sound—Muskoka, ON]

For the short-form census, it's the same penalty as before. The agricultural census—

[Anthony Rota, Liberal, Nipissing—Timiskaming, ON]

So you're willing to throw people in jail for not filling out a short-form census but not a long-form census. I don't see the logic here. And agricultural farmers you're willing to throw in jail, but not the people who won't fill out the long-form census. I fail to see the logic and the connection. You're getting rid of something that's useful when you really should be looking at the penalty, which is—

[Tony Clement, Conservative, Parry Sound—Muskoka, ON]

I know that you and I disagree, but for the agricultural census, obviously I rely on the advice of the Minister of Agriculture. For the short-form census, as we've said, there are eight questions or thereabouts. Basic information is asked, and we do require that every Canadian household that receives that fills it out.

Our problem is not only with the threat of jail time, it's the intrusiveness of the questions. I think I've made that pretty clear here.

[The Chair, Conservative, Michael Chong]

Thank you, Minister, and thank you, Mr. Rota.

We're going to go now to Monsieur Petit.

[Daniel Petit, Conservative, Charlesbourg—Haute-Saint-Charles, QC]

Thank you, Mr. Chair.

Good morning, Mr. Minister. Thank you for coming.

I would like to ask you a general question. I am from Quebec and, according to the last survey, we are the most reluctant province to answer under threat of jail. More than 62% of the people I am representing do not want to answer these types of questions under threat of jail or fine. That is what we are experiencing in my riding. Quebec is the most reluctant province.

Mr. Minister, this is what I would like to know. The media made a lot of noise when our government decided to make some of the mandatory questions voluntary. As I said in the preamble, I do not think that Canadians must be forced, under threat of fine or jail, to disclose private and personal information.

First, I would like to thank you for having the courage to stand up for privacy. It is very important. This is the first government in 30 years to protect privacy in this way.

Could you tell Canadians why the government made this decision and why you came to this decision, other than the fact that the surveys show that Quebec does not want the so-called mandatory system?

[Tony Clement, Conservative, Parry Sound—Muskoka, ON]

Thank you for your question.

Actually, the short-form census is mandatory and is distributed to all Canadian households. However, the 2006 long-form census, for example, was a 40-page questionnaire with 66 very detailed questions on language, education, housing, ethnicity, religion, citizenship and immigration, income, activities, and so on. Section 31 of the Statistics Act indicates that every person who refuses or neglects to provide any information is guilty of an offence and liable on summary conviction to a fine not exceeding $500 or to imprisonment for a term not exceeding three months, or to both. It is a very harsh measure.

So, we decided that it was important to strike a fair balance between the needs of private companies, institutions, the provinces and territories to obtain more information, and the protection of Canadians. This way, if Canadians decide that this information is too personal, they have an option.

[Daniel Petit, Conservative, Charlesbourg—Haute-Saint-Charles, QC]

Mr. Minister, you talked about the short form and long form in terms of balance. You want to remove the obligation to fill out the long form and the threat of jail or of being fined.

As the parliamentary secretary to the Minister of Justice, I have a hard time with that. The opponents of our decision do not want to send real criminals to prison, but now they would like to make criminals out of honest people. That's exactly what they want to do by taking this stand. They want to keep jail sentences and fines the same if someone does not answer the long questionnaire. So, here in the House, we are imposing a more severe penalty on an honest citizen than on the real criminals.

Could you explain how you would be achieving the balance between the long form and the short form?

[Tony Clement, Conservative, Parry Sound—Muskoka, ON]

You are right. We actually want to find a fair and reasonable balance. That is why, after we decided to change the long questionnaire, I asked Statistics Canada to find a way to obtain the necessary information and protect Canadians at the same time.

[The Chair, Conservative, Michael Chong]

Thank you, Mr. Minister.

Mr. Nadeau.

[Richard Nadeau, Bloc, Gatineau, QC]

Thank you, Mr. Chair.

Good morning, Mr. Minister. Good morning, Mr. Dicerni.

The census issue we are currently debating is truly very disturbing. First, it is important to note that the Conservatives also conducted this type of census in 1986, under the Progressive Conservative Party, and in 1991, under Mr. Mulroney. The census had the same parameters.

As to imprisonment, could we not start with the fact that no one has ever been sent to jail? Why aren't you taking the initiative to simply eliminate this aspect? Rather than creating a state of psychosis or something like that, we would simply no longer talk about it. The fact that you are using this kind of argument to back out is extremely disturbing.

Scientists who make a living by doing studies based on statistics, including those of Statistics Canada, have clearly said that, on a scientific level, we would not get the same results from a voluntary questionnaire as from a mandatory questionnaire. It is not me saying that. It is people who have a Ph.D. and who work in this field. It is not gossip in the street or some political will to avoid frightening Canadians. As far as I know, 95% of the people who receive the long form fill it out. So we are creating a state of psychosis, a non-existent problem.

Perhaps we should first look at the origin of censuses. In modern times, and that is where I would like you to be rather than going back to the dark ages, as we went from absolute monarchies to some forms of democracy, information was needed to help citizens in all areas. The objective was mainly to combat poverty and improve health care conditions. At the time, we called it political mathematics. We now call it statistics. It is very useful for the state, especially for helping the population.

Mr. Minister, by choosing to send the long form to 30% of the population instead of 20%, as part of a voluntary process, you are ready to run the risk of having less data and fewer good results on a scientific level. Comparative elements will no longer be acceptable.

Under these conditions, we must remember that we will no longer provide the same service to Canadians. The City of Gatineau is against your approach, and so is the Province of Quebec. A host of organizations, institutions—parliamentary or otherwise—and citizens' representatives have spoken out against this change, as opposed to the few allies you have mentioned.

Mr. Minister, are you ready to back off, to admit that you followed the wrong path and agree that the current approach should be maintained, based on the comments of people who know this field?

[Tony Clement, Conservative, Parry Sound—Muskoka, ON]

We are seeking to achieve a fair balance between the needs of data users and the burden placed on Canadians. Of course, as I have said today and over the last few days, replacing the mandatory survey with a voluntary survey is a challenge in terms of data accuracy. That is why we asked Statistics Canada's opinion on how to structure the national survey in order to reduce this risk. So, we chose the option of sending the form to more households, to one in three, which represents 33% rather than 20% of the population.

[The Chair, Conservative, Michael Chong]

Thank you, Mr. Minister.

We are now going to the third round, with Mr. Lukiwski.

[Tom Lukiwski, Conservative, Regina—Lumsden—Lake Centre, SK]

Thank you, Chair.

Mr. Minister, I want to go back briefly to an observation Mr. Angus made earlier. He was apparently quite critical of my colleague Mr. Lake, who had held round tables in his constituency informing some of his constituents that there's a threat of imprisonment or a fine if you don't fill out the long-form census. Mr. Angus seemed to be suggesting that most Canadians don't realize this, and so no harm, no foul.

Would that be an accurate commentary, that just because people don't realize that there is a threat of imprisonment or a fine we should just leave things the way they are?

[Tony Clement, Conservative, Parry Sound—Muskoka, ON]

Obviously not, and these issues tend to percolate up the closer and closer we get to census time. They reach a crescendo when the census takers start knocking on doors and doing their follow-ups.

The fines and/or imprisonment are in the legislation. They're found in the Statistics Act, and that is part of the repertoire of inducements and/or threats that are used to get people to fill out the form. So we think that's inappropriate. We think that's government going too far in the pursuit of more and more data.

We acknowledge that data is important for making decisions, but we think we should be searching for a middle way of protecting citizens from the overbearing nature of the state when it comes to these things, and at the same time obtaining usable data that serves the purpose for which the survey is intended. That's the balanced approach we're taking.

[Tom Lukiwski, Conservative, Regina—Lumsden—Lake Centre, SK]

I know we'll have officials from Statistics Canada appearing before this committee later this morning, and I'll be asking them similar questions, but my understanding is that Statistics Canada now has a number of voluntary surveys that they utilize in the collection of information for various purposes. Is that correct?

[Tony Clement, Conservative, Parry Sound—Muskoka, ON]

Absolutely. In fact, that role of StatsCan has grown over the years with more and more demand for more and more information. The government gets to make a decision pursuant to the legislation, and we have decided that in order to obtain the right balance, more of the information required from Canadians will be in the form of voluntary questions, and less will be mandatory. That's simply the decision we have made, and I think it's a fair and reasonable decision to be made.

[Tom Lukiwski, Conservative, Regina—Lumsden—Lake Centre, SK]

In the case of the multitude of surveys that Statistics Canada puts out on a voluntary basis--again, this question would be better served by asking those officials, which I will be doing--I assume that a lot of information comes back on a voluntary basis, otherwise Statistics Canada would probably be lobbying for many of those surveys to be made mandatory. Is that a correct assessment?

[Tony Clement, Conservative, Parry Sound—Muskoka, ON]

I encourage you to ask that question of some of the deponents today as well, but certainly the fact of the matter is that I'm here to say, as the minister responsible for StatsCan, that StatsCan is a highly professional organization. They do their job well not only on mandatory censuses but also on voluntary surveys. I have certainly received a commitment from StatsCan that they will apply the same amount of professionalism, in the case of the national household voluntary long-form survey, as they do in other voluntary surveys.

[Tom Lukiwski, Conservative, Regina—Lumsden—Lake Centre, SK]

Finally, Mr. Minister, I would just ask you this. It seems like an obvious question, given the government's position, but I'd like you to express it verbally to this committee, if you can.

What is your level of confidence that changing the long form to a voluntary basis would elicit the same information necessary for Statistics Canada to do the work they have been doing over the past number of decades?

[Tony Clement, Conservative, Parry Sound—Muskoka, ON]

Well, there's no question, if I can say, that if you don't threaten people with jail and/or fines, you have to be a bit more persuasive in moral suasion, using non-coercive methods, to obtain the same information. I don't want to be on record saying anything other than that, but it can be done and it will be done.

[The Chair, Conservative, Michael Chong]

Thank you very much, Mr. Lukiwski.

Thank you to our two witnesses, the deputy minister and the minister, for appearing.

This meeting is suspended until 10:30.

[The Chair, Conservative, Michael Chong]

Welcome to the 29th meeting of the Standing Committee on Industry, Science and Technology on July 27, 2010. We're here pursuant to Standing Order 108(2) for a study of the long-form portion of the census.

In front of us today we have two former chief statisticians of Statistics Canada, who are appearing as individuals: Mr. Sheikh and Mr. Fellegi. Welcome to you both.

I understand that Mr. Sheikh has with him his lawyer, Madam McIsaac.

You're welcome to advise Mr. Sheikh, but I'd ask that you direct your advice to him directly and not to members of the committee.

Mr. Sheikh, you're free to consult with your lawyer at any time. I'd just ask that you and you alone direct your comments to the committee.

Without further ado, there will be an opening five-minute statement from both Mr. Sheikh and Mr. Fellegi, beginning with Mr. Sheikh.

[Dr. Munir Sheikh, Former employee of Statistics Canada, As an Individual]

Honourable Chair, members of the committee, I want to thank you for giving me this opportunity to appear before the committee today.

I think it will be useful, given the recent discussion of issues related to the census, to briefly describe the census process. This will be the basis for my five-minute remarks.

I'll describe this process with reference to two objectives that Statistics Canada tries to achieve, among others--first, to get as high-quality data as possible; and second, to gather these data at the least cost possible. These are the kinds of approaches that have given Statistics Canada the international reputation it has.

Let me break this entire process into six parts.

The first step is consultations with stakeholders and data users. The 2011 Census Content Consultation Guide was released in July 2007, and the 2011 Census Content Consultation Report was released in July 2008. Statistics Canada received more than 1,200 content-related suggestions from more than 150 organizations and private citizens.

The second step is the development of questions. Based on these consultations, and the need for continuity of historical information, changes from one census to the next are generally quite small. Sixty per cent of the questions asked in the 1971 census still remain. Questions in the 2006 census now dropped are related to unpaid work. Questions added for the 2011 census include commute time, child care support and its costs, and subsidized housing.

The third step is to determine which questions go into the short form and which to the long form. The distinction between the population short and long forms began with the 1971 census, as before that there was just a single questionnaire.

Statistics Canada does not differentiate among these questions on the basis of their importance, as they are all tied to the needs of a variety of users. For example, the head-count question in the short form may be the most important for governments in the distribution of parliamentary seats or federal-provincial transfers. But for a city's transportation planning department, the most important piece of information may be about how people get to work and how much time they spend commuting, a question that is in the long form.

It therefore is not the importance of the questions that determines whether they are in the short or the long form. What determines this division is a cost-efficiency test: how can we get the highest-quality data possible at the least cost?

To the best of my knowledge, those working at Statistics Canada on the 1971 census determined that for reasonably comparable levels of quality, some questions must be asked of the entire population. These ended up in the short form. For other questions, the required quality of information would be achieved by scientifically selecting a representative sample and making it mandatory for the sample to respond; this became the long-form questionnaire.

Naturally, the cost of using one-fifth of the population as a sample, rather than the whole population, reduces the cost of the census significantly, while reducing the response burden substantially as well.

Since 1971 Statistics Canada methodologists have reviewed this distribution of questions between the short and the long forms for every census cycle. This is the only reason why the following question is on the short form--and I quote--“What is the language that this person first learned at home in childhood and still understands?”, and on the long form, “During the week of Sunday May 1 to Saturday May 7, 2011, how many hours did this person spend working for pay or in self-employment?”

The fourth step in the process is for Statistics Canada to submit the proposed questions to the government and for the government to review these proposed questions and tell Statistics Canada what the census content will be.

The fifth step is to collect data. Again, to constantly strive to reduce costs, Statistics Canada moved in a substantive way to Internet data collection in the 2006 census. The May 2009 census test indicated the potential to double this response rate in 2011 to about 40%.

Statistics Canada's per-dwelling census cost is $43.77 in 2009-10 dollars, which compares with the cost of $126.18 in the U.S. and $49.68 in Australia, all in terms of Canadian dollars. Both are countries that we normally compare ourselves to.

The sixth step is to transform raw data into data that is useful for the users and to disseminate it. Dissemination begins about a year after the census, as the large amount of data has to be processed, edited, and checked for accuracy, gaps, and consistency. The data is analyzed thoroughly to understand the results before it is released publicly.

The whole process takes about seven years.

Thank you, Mr. Chair.

[The Chair, Conservative, Michael Chong]

Thank you, Mr. Sheikh.

Mr. Fellegi, you have five minutes for an opening statement.