Official Languages Committee on March 9th, 2017

[The Chair, Liberal, Denis Paradis]

Welcome.

Pursuant to Standing Order 108(3), we will resume our study of the issues related to the enumeration of rights-holders under section 23 of the Canadian Charter of Rights and Freedoms.

We are pleased to welcome this morning two officials from Statistics Canada: Ms. Connie Graziadei, assistant chief statistician, and Ms. Johanne Denis, director general. Also with us is Mr. Larry Shute, from the Department of Industry.

Welcome, everybody.

Ms. Graziadei and Ms. Denis will have about 10 minutes for their presentation, and we will then go around the table to allow committee members to ask questions and make comments.

Please go ahead, Ms. Graziadei.

[Connie Graziadei, Assistant Chief Statistician, Census, Operations and Communications, Statistics Canada]

Thank you very much, Mr. Chair.

I want to thank the committee for inviting us to appear here today.

The most fundamental role of Statistics Canada is to support the democratic process by providing Canadians with high-quality, current, and relevant information on the state of the nation. Comprehensive, objective, and high-quality statistical information provides a solid foundation for informed decisions by you, who are our elected officials, and by businesses, unions, and non-profit organizations, as well as individual Canadians. Statistics Canada produces information that helps Canadians better understand our country, population, resources, economy, society, and culture.

As the national statistical agency, we take pride in our work and take our professional responsibilities very seriously. It is our job to understand the information needs of the nation and to address these requirements in a scientific, neutral, efficient, and effective way. We are committed to protecting the confidentiality of all information entrusted to us and to ensuring that the information we produce is timely and relevant to Canadians.

Today we are faced with continuous and growing demands for more detailed and more timely information. This is true not only for the census program but also for our ongoing social, economic, and environment programs as well. We have a long-standing history and world-leading reputation in using innovative methodological approaches and instruments to address these growing information needs.

The demand for information to be collected on the census program is not a new phenomenon. My colleague will explain to you the rigorous consultation program that we undertake to ensure we remain relevant and respond to emerging priorities.

While the census is incredibly successful and provides a richness of information at the community level, it is not necessarily the right instrument or approach to deliver high-quality information for specific topics. Some questions that are specifically designed to capture detailed information from subpopulations, for example, are better asked on a survey or captured through existing administrative sources rather than from the whole of the population.

Our social statistics program can respond to information needs faster, and we can use our analytical capacity to model and link data sources to produce information at lower levels of geography.

Our comprehensive social statistics program also includes a robust cost recovery program. One example of our cost recovery program is the use of post-censal surveys, such as the aboriginal peoples survey that is currently in the field or the survey on the vitality of official-language minorities that was conducted following the 2006 census. The post-censal survey program exemplifies Statistics Canada's partnership with other federal departments or a consortium of departments that provide funding to the agency to address specific emerging priorities.

Before I describe the process by which we obtain approval for our questions that are asked on the census program, I would now like to turn it over to my colleague, Johanne Denis, to describe our comprehensive census content consultation process that we will embark on later this year and to speak on the subject of collecting the additional language data.

[Johanne Denis, Director General, Census Subject Matter, Social and Demographic Statistics, Statistics Canada]

Hello.

The framework for determining the content of the census program is based on a proven process that draws on experience from previous Canada census programs, the United Nations' census principles, and approaches developed in other countries. This framework determines the priorities in the census content, in keeping with users' information needs, the burden on the respondent, privacy issues, and operational considerations. This framework is used and rigorously applied during the consultation process with data users and partners.

Starting in fall 2017, Statistics Canada will be implementing an official public consultation process. Following the consultation and the assessment of needs according to the framework, any changes proposed to the census content will be rigorously evaluated, including qualitative and quantitative testing, at the high quality standards of Statistics Canada. This rigorous and scientific approach will make it possible among other things to evaluate the effect of new content on the existing content

As to gathering additional linguistic data, Mr. Jean-Pierre Corbeil stated in his appearance on February 14, 2017, that efforts that had been made in the 1993 and 1998 census tests to assess the feasibility of including a question on the language of instruction of people aged 15 and over. The results of these tests showed that it was extremely difficult to obtain that information, and that there were major gaps and inconsistencies.

I will now briefly outline two approaches that could replace the census and that could be used to collect data about rights-holders. As the national statistics agency, it is our responsibility to respond to information needs by recommending the appropriate tools to precisely and accurately report on rights-holders.

First, similar to what was done in 2006, a survey could be conducted in 2021 and the answers to the 2021 population census could be used to select a sample of individuals belonging to official language minorities. Funding would be needed for this survey and the sample size would be determined by the geographical level required. Statistical methods and techniques such as modelling and oversampling could be used to estimate the population of rights-holders for small geographical areas, such as areas served by school boards.

As to the second approach, Statistics Canada has for a number of years collected a wide range of data from across the country, in collaboration with the provinces and territories. Each year, the agency receives data from every province and territory about kindergarten, primary and high school enrolment. If the provinces were able to collect standardized data on the language of instruction of parents, brothers and sisters, and provided that data to Statistics Canada, it would be possible to enumerate rights-holders using administrative data.

Statistics Canada has already begun discussions on the enumeration of rights-holders with board members of the Fédération des communautés francophones et acadienne du Canada. In the interest of upholding the spirit and implementing section 41 of the Official Languages Act, Statistics Canada is committed to developing and implementing the best strategy for collecting data on the subject.

[Connie Graziadei]

As you've just heard, the census consultation and testing process for the 2021 census will begin later this year. The process by which the questions for the census are approved is determined by legislation. Section 21 of the Statistics Act states:

The Governor in Council shall, by order, prescribe the questions to be asked in any census taken by Statistics Canada....

To ensure the Canadian public are informed of this decision, the act also requires that the questions be published in the Canada Gazette no later than 30 days after the decision.

For the 2021 census program, Statistics Canada plans to present its recommendations to cabinet in fall/winter 2019, with the goal of having the questions published in the Canada Gazette no later than spring 2020. This timeline is required to ensure all systems and processes are properly adjusted, tested, and finalized for collection to begin in early 2021.

ln closing, Mr. Chair, please let me assure members that Statistics Canada takes its professional responsibilities very seriously, and we remain very committed to producing the highest-quality information using the most innovative methodological approaches at our disposal.

My colleague and I would be happy to address any questions you may have. Thank you.

[The Chair, Liberal, Denis Paradis]

Thank you very much for your presentation.

We'll now go to questions and comments from the members.

We will begin with Mr. John Nater.

[John Nater, Conservative, Perth—Wellington, ON]

Thank you, Mr. Chair, and thank you to our witnesses for joining us this morning to discuss this issue.

I think at the root of all this is the issue of identifying rights holders and quantifying the number. It sounds fairly simply, but as we've heard from past witnesses, it's obviously not as simple as asking someone, “Are you or are you not a rights holder?” The terminology doesn't quite jibe with individuals when they're being questioned. The way we ask these questions in a census and/or survey is important.

Now in your presentation, madam—I'm not going to be able to pronounce your name correctly—

[Connie Graziadei]

That's okay.

[John Nater, Conservative, Perth—Wellington, ON]

—you mentioned that the census isn't always the right tool to find the information we're gathering.

The challenge I have with that statement is that when we're talking about rights holders, we're talking about an entire population. We have a sample size of 35 million Canadians. Of those, any one could potentially be a rights holder. By using an alternative sample size, even with supersampling, we're not really quantifying the number of rights holders in a given community.

I was talking to a colleague from Peace River, Alberta, who has a number of small bilingual communities in northern Alberta, with bilingual schools there. Even with supersampling, we're not going to necessarily identify where these rights holders are, to provide that information for education.

Could you elaborate on this problem a little, in terms of identifying the right tool to find this information?

[Johanne Denis]

Do you want me to answer?

[Connie Graziadei]

Sure, go ahead.

[Johanne Denis]

On the idea of a survey in which we target using the census, we could do a post-censal survey similar to the survey on the vitality of official-language minorities. We could use techniques such as oversampling—you're right that we can go to a certain level—but we have other capacities as well. We can use a modelling approach to model at the small area level. As a scientific organization, we have used these techniques on many occasions. This is something with potential. We will need to test it and look at it, but it is something that is feasible.

On top of that, there were my suggestions about administrative data. Administrative data allows you to cover 100% of the population as well. This is another avenue worth looking at.

[John Nater, Conservative, Perth—Wellington, ON]

Okay.

The final decision of which questions are and are not included in the census comes from the Governor in Council. Perhaps Mr. Shute, from Industry Canada, could comment on that.

Who makes the final recommendation to the Governor in Council, to cabinet? Is it Industry Canada that has the final say, or is it Stats Canada that provides a final recommendation to the minister and through that mechanism to cabinet?

[Larry Shute, Deputy Director General, Economic Research and Policy Analysis Branch, Strategic Policy Sector, Department of Industry]

Thank you for your question.

The recommendations on the questions for the census content rely most importantly on the work that Statistics Canada does to consult with Canadians. They make a recommendation to the minister, who then makes a recommendation to the Governor in Council.

[John Nater, Conservative, Perth—Wellington, ON]

Okay. Do you know of examples in which disagreements have occurred in those recommendations?

[Larry Shute]

Do you mean between Statistics Canada and the minister?

[John Nater, Conservative, Perth—Wellington, ON]

Yes.

[Larry Shute]

There are none, in my experience.

[John Nater, Conservative, Perth—Wellington, ON]

Typically, then, what is recommended is what is presented?

[Larry Shute]

That's correct.

[John Nater, Conservative, Perth—Wellington, ON]

Okay.

We talked a little bit about the survey on the vitality of language in minority communities, from 2006. There's no question that some important data came to us through that study. Concerning when this data is being used in courts, we had a witness, Mark Power, state that you really can't use this information in court, that it doesn't hold up.

How would you react to that? What would be your opinion on that criticism, I would say, of this particular survey?

[Johanne Denis]

It's not for me to say at this stage. We're embarking on census consultations, and certainly we hear you. We will consult as well with the organization, as we do already. As I said, Mr. Corbeil is in close communication with la Fédération. This will be part of the census consultation, and we will take your concern seriously and will look at potential ways to do additional testing. You need to understand, however, that we cannot stop looking in parallel at various options. As a scientific and rigorous statistical organization, we take this question very seriously, and work continues on that front.

[John Nater, Conservative, Perth—Wellington, ON]

Going forward, then, for the next census in 2021, have there been any guidelines released by Stats Canada in terms of how many questions are envisioned overall in the short-form census and in the long-form census?

[Connie Graziadei]

That's part of the process we are embarking on now. When we're doing the consultations, we are not only looking for information needs but also at information comparability over time and the importance of that to the data users.

Obviously there are financial constraints, but probably more importantly there are also constraints of burden on Canadians, and we have put a tremendous amount of effort over the last 10 years in trying to reduce the burden on Canadians by replacing information questions with a very highly reliable source of information. An example in the 2016 census is that we didn't ask the 14 questions on income that had been previously asked, because we had a very reliable source of information.

We go through the consultation process, and it really is at the end of the consultation process that we determine what needs to be asked of 100% of the population versus what we can ask of a sample of the population, and then we use estimation in a very scientific and professional way to produce reliable information at lower levels of geography.

[The Chair, Liberal, Denis Paradis]

Thank you very much.

Mr. Samson, you have the floor.

[Darrell Samson, Liberal, Sackville—Preston—Chezzetcook, NS]

Thank you, Mr. Chair.

Thank you very much for your presentation, ladies.

I have just six minutes so the questions and answers will have to be very quick.

My first comment is general in nature. I hope the feeling I had after your presentation was not warranted. I had the feeling in my heart, in my guts, that you were here to tell us that adding a few questions might not be the best way of gathering information. I hope that is not true, but I will leave that in your hands.

I was very interested in the way you described the context at the beginning. You are experts, and we do not question that. We want to inform Canadians and you find innovative ways of doing that. That is all good.

Mr. Corbeil appeared before us and stated that the census does not include questions pertaining to the subject of section 23(1)(b) or paragraph 23(2) of the Charter.

Do you share his view?