Industry Committee on Aug. 27th, 2010

[Pierre Noreau]

My opinion is based on the body of research carried out. I do a lot of quantitative analysis. The response rate for surveys we use and administer in ordinary conditions is steadily declining. A few years ago, approximately 75% of the population answered questions for an ordinary survey. Today, a response rate of 50% is considered to be good. And yet, it is clear that an entire segment of the population is underrepresented. This cannot happen with the census.

[Mauril Bélanger, Liberal, Ottawa—Vanier, ON]

I will come back to that later.

In your view, does Canada have international obligations with respect to information sharing, which it would not be able to fulfill if the long form was voluntary?

[Pierre Noreau]

I do not know the extent of Canada's commitments to the OECD with regard to providing specific data about its population. However, I do see how this would be useful. It is not just a question of legality or obligation. It is also Canada's credibility as a modern society that is at stake. It deals with other societies that make the effort to study the reality of their own populations. It is a question of international credibility.

[Mauril Bélanger, Liberal, Ottawa—Vanier, ON]

Ms. Taillon, earlier this week there was an article published in the journal Nature, written by two leading statisticians in the United States—actually, one of whom came from Canada, Mr. Fienberg, who grew up in Toronto and was vice-president of York University; and Mr. Prewitt, a former director of the U.S. Census Bureau. They make a rather strong affirmation in there that this is a short-sighted decision that will end up costing us billions more, which is somewhat along the lines of what you said.

They also say that detailed, reliable data is needed for everything from determining how many hospitals are needed to tracking how poverty and prosperity relate to health or education. They say,Census data provide the gold standard against which all other studies on such issues can be corrected and judged.

This morning we had Mr. Mel Cappe, who is heading up a fairly well-respected research organization in Canada, and he used to be Clerk of the Privy Council. In response to a question from one of the government members, he brought up a point that has not been explored at great length, and unfortunately we've not been able to get the Canadian Nurses Association before us, although they want to appear. It has to do with the social determinants of health.

In your position as chair of your council, could you elaborate on the usefulness of the long-form census data in terms of social determinants of health and what that all means?

[Peggy Taillon]

Yes, I'd be happy to.

Just to give you a bit of background, before I came to CCSD, I actually was the senior vice-president at the Ottawa Hospital, one of the largest academic health science centres in the country. So I often call myself a “recovering hospital administrator”, and still in recovery.

But I would say, from the perspective of the social determinants of health, the information in the mandatory long form is really, again, as I mentioned, kind of a navigation tool. It focuses on those aspects that determine our health, such as income levels, whether you are a single parent, whether you have supports such as child care so that you can get to work. All those things are taken into consideration, because we all know that all of them determine our path through the health care system.

It also helps us project out forward. We all know that we have this aging population in Canada. It's not enough just to know that we have an aging population; we also have to understand where to find them. Where do we need to focus our resources? Are they in urban areas? Are they moving back to rural areas? The long form helps us do that. It's a fundamental tool for that work.

[Mauril Bélanger, Liberal, Ottawa—Vanier, ON]

Since the Government of Canada has the responsibility for aboriginal health issues, contrary to the provincial responsibility in non-aboriginal populations, would the same apply, or would it even be more compulsory or more necessary that the Government of Canada have very accurate information vis-à-vis its aboriginal population?

[Peggy Taillon]

As the government responsible for the provision of health and social services to aboriginal Canadians, it absolutely has the responsibility for it. If you survey or speak to aboriginal leaders across the country, they'll tell you that the data that we have right now on aboriginal Canadians isn't the best. So a lot of work needs to be done there. But the mandatory nature of the census actually gives the leadership in aboriginal communities a lever to help them work with their local communities to get that data from people who typically would not fill out a survey.

[The Chair, Conservative, Michael Chong]

Thank you.

Monsieur Bouchard.

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

Welcome madam and gentlemen. Thank you for appearing before us.

My first question is for Mr. Noreau.

Does the data, the information collected with the mandatory long form, provide a precise snapshot of society and the general population?

[Pierre Noreau]

I believe that the response rate for the questionnaire is presently 97%. In fact, it is this rate that assures researchers and governments that the data is reliable. That is the case for any study. In terms of data certainty, the most important element is the percentage of people who participate in the survey. The closer this percentage comes to the total number of individuals in the population studied, the more accurate the snapshot and the easier it is to use the data for other, much more refined studies. It results in a much more nuanced understanding of our reality.

The participation rate is important but the number of participants is also quite crucial. For example, if you need information about the situation of young parents under 24 years of age—and God knows that young families are a reality—a sufficient number of young people must participate in the survey in order to obtain the data required to establish family policies, for example. That is the case for all areas. The greater the number of respondents, the more accurate the data, the more nuanced the analysis of the data, and the more your public policies are in tune with reality.

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

In short, the mandatory long form provides an accurate snapshot.

[Pierre Noreau]

It is more specific, more reliable. It can be very reliable given that the response rate is adequate and that it covers practically all the characteristics of the population being studied.

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

All right.

How would you describe a snapshot of a society and its population based on data collected using a voluntary questionnaire?

[Pierre Noreau]

In that case, the snapshot would be much more blurry. As I was saying earlier, some segments of the population do not respond or do not readily participate in surveys of this kind when they are voluntary. Consequently, those people are underrepresented in the sample, in the part of the population to be studied. Therefore, there will not be enough data about them to really take their situation into account when the time comes to establish public policies.

This applies to public policies. However, as mentioned earlier, it also applies to determining needs. Should hospitals be built? Where should a school and, in future, shopping centres be built ? All these questions require very detailed information about the population and, therefore, a very accurate measurement. In order to be accurate, the participation rate must be high and, therefore, the questionnaire must be mandatory. This is all the more true given that other surveys we conduct cannot be made mandatory. That is because at this point we generally cannot force people to participate in an existing survey and we need this one. The number one survey, on which all the others are based, must be as complete as possible. It is truly the foundation for all research on Canada's population.

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

Mr. Noreau, with regard to the statistics, will the decision to make completion of the long form voluntary rather than mandatory diminish the scientific value of the data collected?

[Pierre Noreau]

Yes, that is obvious. It will reduce their reliability. In the long term it is going to be a problem for reasons that have not yet been explained. When a survey such as the census is carried out in a specific year—2011, for example—we obtain a snapshot of Canadian society at that point in time. When a number of subsequent censuses are conducted, it is as though we have a movie about our society. It is not just a snapshot. We can see how our society is evolving.

If you wish to plan future public policies, you have to see this change. You can only measure it by conducting the survey in a consistent manner. For this reason, it must continue to be mandatory. It is the only way to reliably compare the results of the next census with those of previous ones.

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

My next question is for Ms. Taillon.

You talked about taxpayers' money. When you conduct a census, taxpayers' money is spent. You are aware that this change represents an additional expense of $30 million for the government. Moving from a mandatory questionnaire to a voluntary one represents an important change.

Do you think that spending $30 million on this change represents value for money for citizens?

[Peggy Taillon]

Absolutely not, because it's going to cost us more and we're going to get less accurate data. If you follow the line of thinking I mentioned earlier, you're going to get less responsive services on the ground and communities aren't going to be as well served as they were in the past. Some of the implications of this we'll see over decades, and it compounds. So I'd say absolutely not.

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

Mr. Noreau, how does the mandatory form contribute to increasing knowledge in the social sciences?

[Pierre Noreau]

It is the basis for calibrating almost all research conducted in a vast number of social sciences, especially economic analysis, sociology and political science, three areas with which I am particularly familiar.

Every time we conduct a survey, we must assess its value by comparing it to census data. For example, it is possible for housewives to be overrepresented in a survey. In fact, with a telephone survey, the people most likely to respond are those who are at home. They are often women. Consequently, in most of our surveys, there is overrepresentation of housewives and underrepresentation of youth, who cannot be reached at home during the day, in the evening, or even late at night.

The only way to take into account the fact that our survey has flaws is to recalibrate the survey based on census data. We use the latter to correct our sample. That is why the census data must be the most reliable. We use the data to correct all our other surveys. In most areas, particularly the social sciences and humanities, it is vital to have a completely reliable survey. In fact, all the others have flaws and we truly need the census data to correct them. That is the case for research.

If census data were to become less reliable one day, we would have serious problems with the accuracy of our research in the six-year period after the census. It would be a very big problem for the social sciences and humanities.

[The Chair, Conservative, Michael Chong]

Merci, monsieur Noreau.

Mr. Lake.

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

Thank you, Mr. Chair; and to all the witnesses today, thank you for coming.

Actually, before I ask a question, maybe I'll just see if Mr. McFarlane wants to weigh in on anything he has heard so far.

I'll give you the opportunity, if you have anything to add or comment on from what you've heard from the other witnesses.

[Lawrie McFarlane]

Thank you. I have two or three quick points.

We already have difficulty with the existing census and getting responses from native groups. Some native groups refuse to comply, in part because they don't like the element of compulsion.

The United States carried out a voluntary survey some years back at the request of Congress. What they found was that response rates dropped about 20 percentage points, meaning that they were getting fewer than they would have gotten but still a large number.

It was going to cost more, by all means, to do a voluntary process, but let me put these numbers in context.

There are roughly 14 million families in Canada. Let's assume 10% of them are poor: 1.4 million poor families. The response rate in the United States for poor groups was 20%. In other words, for the voluntary census in the United States, 20% of black, urban, poor Americans responded. If we got a 20% response rate in our groups of poor families, we would get 280,000 responses. That's a huge sample size. There is clearly room here for us to have a voluntary survey in which admittedly the number of responders will fall, but because we're starting out with such a massive base—this is the whole of the population we're surveying—we would still end up, I think, with a very significant response.

Although I may have misheard, I thought I heard Monsieur Noreau say we shouldn't be forcing people. If that's the case, there is no dispute here. Nobody that I know of thinks we shouldn't be trying to get this information. The question comes down to the degree of compulsion.

Thank you.

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

Ms. Taillon, you had made a comment and I just want to clarify. You talked about the aging population. So to understand the aging population, not only do we need to know who they are but where to find them. I think you were saying not only do we need to know their date of birth, but we also need to know their address, where they are distributed around the country. Is that accurate?

I think you also touched on school location in your opening comments, and it's the same thing there. So it's more than just knowing how many people there are at whatever age; we need to know where they are located as well.

[Peggy Taillon]

We need to understand it so that we can put better services on the ground and understand best where to put geriatric services versus day care, and so on, when we have a better sense of where they are specifically located.

It's the same for new Canadians. If we're targeting a particular area that we know is struggling and has lower incomes in Canada, it's good to know that, but it's even better to know, are there a lot of new Canadians? Are there specific new Canadian populations that have moved into that area? Do we need to tailor services in that area to Somali Canadians, for example?

That is the beauty of it, because it really takes it down to the neighbourhood level.