National Statistics Council Statement on Census

Statement issued yesterday:

Welcoming the Changes Announced on August 11th

The National Statistics Council, the senior, external advisory group appointed by the government of Canada to advise the Chief Statistician, has noted the Government’s announcement that it intends “to remove threats of jail time for persons refusing to fill out the census” and its recognition that the voluntary National Household Survey will not meet the requirements for robust and accurate small-area data (e.g. data related to the use of Canada’s Official Languages) that can only be provided through a mandatory instrument.
The Council supports these developments, and is gratified that the government has acted in a manner consistent with suggestions about removing jail time outlined in the Council’s July 25th statement Seeking Solutions. International experience with mandatory censuses and surveys has shown that success can be achieved without the threat of jail.

Resolving the Issues by Listening to Canadians

While the initial decision of the government was taken without public consultation, the debate and discussion since the decision was announced have been illuminating. Many groups have explained the importance of the long form for their activities. They have brought to the attention of the government many facts that the government may not have known given the initial lack of consultation.
The public debate and the testimony at the Standing Committee, as well as the information released by the government yesterday in response to the order of the Standing Committee, have made the situation clearer and have demonstrated that:

All expert statistical advice, as well as the results of the recent U.S. survey experiment, concurs that a voluntary survey will not be able to fulfill the fundamental needs of a national statistical system in terms of accurate data for small-area needs, or for the benchmarking needed to ensure that other Statistics Canada surveys accurately represent the Canadian population.

The cost of changing the long-form part of the census into a survey is substantial — $30 million or more — while yielding less accurate or usable information.

While there has been recent discussion that small-area data users such as public health planners, voluntary associations, towns, highway and transportation designers, and many more should pay for their own surveys instead of relying on the census, this is not possible. For the same reason, non-response bias, that a voluntary replacement for the long-form census will not yield accurate results with a 50% – 70% response rate, private surveys will be totally unacceptable with their 10% – 15% response rates. (Again, their low response rates can be adjusted to be more accurate reflections of the whole population, but that requires the benchmarking from the long-form which will no longer be available.)

The Government’s Announcement Can be a Path to a Widely-embraced Resolution

The government’s announcements provide a welcome point of departure for an urgently needed fresh look at the issues and the crafting of additional initiatives that will protect and enhance the Canadian statistical system, ensures privacy, serves the Canadian public better, and reduces costs.

In the short term, we believe that there is an opportunity by making the National Household Survey compulsory to:

Save significant sums of money.

Reduce respondent burden on Canadians as the number of people asked to complete the long-form will be reduced from 30% of the population to 20%.

Provide the accurate benchmarking information needed to ensure that Stats Can and other data users – public, voluntary or private – can gather subsequent information.

Give vital information needed for small area or neighbourhood uses, including housing, health and transportation planning.
The National Statistics Council is eager to work with the government and with Statistics Canada on an urgent basis to achieve these important short-term goals.

Looking to the Future

The National Statistics Council believes that there are important ways in which the census and the statistical system can be improved for all Canadians. While these are not subject to the same time pressure as the current decisions about census content and form, they should be addressed in the coming year.

The National Statistics Council recommends that over the next year:

1.Statistics Canada consult widely and examine the process for deciding what questions merit inclusion in each census. While a suggested set of criteria was set out in the Council’s Seeking Solutions, the important goal is to make the process transparent and the need clearly demonstrable.

2. The opportunity afforded by amending the Statistics Act be used to include provisions related to the UN Fundamental Principles of Official Statistics, a code adopted by the UN and to which Canada has already formally adhered.

3.Statistics Canada examine the respondent burden placed on Canada’s farmers by the Census of Agriculture and other agricultural surveys from Statistics Canada. This review would look at collection techniques, search for ways to reduce the volume of questions, and consult with farmers about their need for data.

The National Statistics Council stands ready to assist the Minister and Government in using the recent focus of public interest in the census to enhance Canada’s system of national statistics so that it meets the contemporary and emerging needs of Canadians.

Ian McKinnon,

Chair, The National Statistics Council


  • The government will do everything it can not to have a census.

    The reason they are canceling it is to hide the massive economic trama.

    The next census would show that the official line that the economy is getting better is false

  • It is none of the government’s, or the National Statistics Council’s, or the United Way’s beeswax how many bedrooms I have in my home, what time I go to work, what my religion is . . . etc. etc. etc

  • In Canada there is a polarization on the subject of the mandatory census. People “for ” the mandatory census speak of the need for the data and it’s value in formulating smart government policy. People “against” the mandatory census talk of their rights and coercion, and does a group have the right to demand an individual answer questions about their life against their will, under threat of fine and jail.
    So what’s more important, the group/state/country/collective or the rights of the individual? Hasn’t this question been played out countless times over the centuries around the world? How was this question settled? I guess it depends which country you live in. Some countries certainly protect the rights of the individual over the right of the group a lot better than others.
    If you study sociology or anthropology or philosophy, it is easy to identify these concepts in other populations/countries/ historical periods, but every population and individual is blind to these concepts in their own group. What do you do when the truth of your group conflicts with the truth of an individual?
    “StatsCan does great work for Canada.” (Truth for our group/Country/population.)
    “I’ll ask the economists, politicians, and government agencies to keep their unwanted, uncomfortable questions off of me.” (Truth for the single mother who feels uncomfortable with the questions the census informs her she must answer, and refuses to answer)
    Hence this debate, because the existence of people who don’t answer the census questions, or answer them incorrectly, or write “object to question” for the reply poses an inconvenient truth for the group (“us Canadians”).
    In fact, if you look at census data from all countries, including Canada, there are a significant number of respondents who mark some answers as “object to question”. Many others answer some questions in obviously incorrect ways which are understood but are not counted (eg. marking Jedi as their religion). It’s not easily known how many people purposely answer incorrectly and have these incorrect answers erroneously counted as accurate by Statistics Canada. Statistical agencies around the world don’t like to talk about this problem, because it skews the data and puts the validity of all their work at risk.
    Now that the issue has been politicized, we even have some people who are “for” the mandatory census saying that people who object to the questions don’t really exist. I’d be careful with that line of thought.
    Everyone knows the census provides valuable data. No one is saying the data is not. You won’t win the argument by listing the hundreds of groups and organizations who want and need this data, unless you deal with why coercion of individuals to get this data is acceptable in this case.

    Is the following acceptable?

    Single Mother: “I feel uncomfortable with you and your questions. Please stop contacting me and leave me alone. I feel you are harassing me and I’m asking you to please stop. I know my rights.”

    StatsCan worker: “We don’t care how you feel, you must answer our questions about your life. I am doing nothing wrong, this is my job. You are the one breaking the law and we can fine you or put you in jail for your behaviour. We will keep contacting you. We will keep coming by your home. We have your name and we have taken a description of you. We want much more information than you have provided and you are being uncooperative. There is nothing you can do about this. You are wrong and you do not have any rights in this situation.” (Truth for the statscan worker who has a quota to reach, and is dealing with uncooperative, unhelpful, anti-government people who are purposely not answering these mandatory census questions properly.)

    The standard in the western world for data collection from individuals is informed consent, (with the one and only exception of the mandatory census). Maybe it’s time to look at this issue and have a real debate about the issues.

    A really long list of the people and groups who use and need this data is also a list of people who are for coercion and don’t believe in informed consent.

  • Thank you Chantal for your thoughtful comments.

    I have recently been selected by Stats Canada for the Labor Survey, which I have refused to participate in for various personal reasons.

    It seems that the issue of voluntarily or being forced to provide your personal information to the government has unfortunately devolved to a political discussion divided upon party lines. For me, refusing to participate has nothing to do with this being a Conservative initiative or one opposed by the Liberals. In fact, I could care less what either party thinks about the issue and am only concerned about my rights and freedom.

    I recognize that the information obtained is used by various parties, though after much research I only have a very general idea of who uses the information and for what purpose. I am going to assume it is being used for the greater good.

    After much reading of the subject, including blogs and discussion groups, I see a large number of fellow citizens who criticize the goverment and/or people who refuse to participate in the survey and feel that it should be mandatory. I also see a great deal of people who knowingly provide false information in order to satisfy the mandatory requirements of the survey and to get Stats Canada off their back.

    I am phoned everyday, sometimes twice a day, and have told Stats Canada on each occassion that I will not participate and am quite prepared for them to charge me under the act. They tell me they will phone me everyday for the next six months and that I can expect their enumerators to appear at my door.

    It seems to me, with the amount of people who feel that the mandatory survey is a requirement in Canada, and who feel that people should be imprisoned for three months for not participating that Stats Canada has a large sampling of the population who would voluntarily participate. Why they spend 100’s and 1000’s of dollars pursuing people who object to providing their personal information to the government, or worse, taking information that is patently false in order to get Stats Canada off their back, makes no sense to me.

Leave a Reply

Your email address will not be published. Required fields are marked *