Overview of the study

The study consisted of five steps:

Step 1

The first step was to compile and review a large number of existing sources of information on census-taking methods in other countries. These sources included several international organizations, including the United Nations Statistics Division (UNSD), the United Nations Economic Commission for Europe (UNECE), and the related Conference of European Statisticians (CES). In addition, a review was conducted of documentation available from numerous national statistics institutes, including the United States (U.S.) Census Bureau, the United Kingdom (U.K.) Office for National Statistics, the Institut national de la statistique et des études économiques (France), the Swiss Federal Statistical Office, Statistics New Zealand and many others.

One of the most relevant sources of information on international census-taking methods was the UNECE. The UNECE is one of five regional commissions of the United Nations (UN)Footnote 1 and currently has 56 member countries, including Canada, the United States, and some countries from Asia. Other non-member countries, such as Australia, often participate in its work. Sources of information from the other four regional commissions were also examined, but very little additional information of relevance was found. Because of their comparable levels of social and economic development, as well as the maturity of their statistical systems, the member states of the UNECE, supplemented by Australia and New Zealand, provide by far the best comparison group.

Step 2

In the second step, internationally recognized principles and recommendations for census-taking were taken into account. In this regard, the fundamental document is Principles and Recommendations for Population and Housing Censuses, Revision 2 (United Nations 2008). This document is intended for use by all member states of the UN in planning the 2010 round of censuses. It describes principles and makes recommendations concerning operational aspects of population and housing censuses, the topics to be included in population and housing censuses, and census products and data utilization. In parallel with the development of the UN document, the UNECE, in cooperation with the Statistical Office of the European Communities (EUROSTAT), produced the document Conference of European Statisticians Recommendations for the 2010 Censuses of Population and Housing (UNECE 2006). The UNECE document makes recommendations for countries within the UNECE region in the areas of census methodology, technology, and population and housing topics.

The UN defines a population census as "…the total process of collecting, compiling, evaluating, analysing and publishing or otherwise disseminating demographic, economic and social data pertaining, at a specified time, to all persons in a country or in a well-delimited part of a country" (United Nations 2008, page 7). A similar definition is provided for a housing census, pertaining "…to all living quarters and occupants thereof…" In the remainder of this document, the terms 'census' or 'census of population' will mean both a population and a housing census.

The UN identifies four essential features of population and housing censuses (Ibid., page 8):

  1. Individual enumeration – Each individual and each set of living quarters is enumerated separately and the characteristics thereof are recorded separately. Only in this way can the data on the various characteristics be cross-classified. This requirement can be met by collection of information in the field, by the use of information contained in appropriate administrative registers, or by combinations of these methods.
  2. Universality within a defined territory – The census should cover a precisely defined territory, for example, the entire country or a well-delimited part of it. Every person present or residing in the territory and every set of living quarters irrespective of type should be included. This does not preclude the use of sampling techniques for obtaining data on specified characteristics. Such universality provides the basis for producing data for small geographic areas and small subgroups of the population.
  3. Simultaneity – Each person and each set of living quarters should be enumerated at the same well-defined point in time, and the data collected should refer to a well-defined reference period. This does not preclude the possibility of having different reference periods for different data items (for example, employment status may refer to the week prior to Census Day while income may refer to the previous calendar year).
  4. Defined periodicity – Censuses should be taken at regular intervals so that comparable information is made available in a fixed sequence. The UN recommends that a national census be taken at least every 10 years, and notes that national and international comparisons are of greater value when countries undertake a census in years ending in '0' or at a time as close as possible to this, although it recognizes that national factors such as legal, administrative, financial and other considerations should take precedence in fixing a census date.

To this list the UNECE adds a fifth feature (UNECE 2006, page 8).

  1. Small-area data – The census should produce data on the number and characteristics of the population and living quarters related to the smallest geographic areas of the country, and to small population groups of interest, consistent with protecting individual confidentiality.

Step 3

In the third step, on the basis of the literature review, the variety of census-taking approaches was classified as follows:Footnote 2

  1. Traditional census
  2. Census employing existing administrative registers:
    • Register plus complete enumeration
    • Register plus survey collection
    • Register plus complete enumeration plus survey collection
    • Register-based only census
  3. Traditional census with yearly updates of characteristics
  4. Rolling census

Within each of these approaches, there exists a wide variety of more specific methodologies. For example, with any of the approaches involving collection of census or survey information directly from respondents (all except 2.(d)), the data may be collected by enumerators or through self-completion of a questionnaire by the household. Questionnaires may be mailed out using an address list compiled in advance or dropped off by enumerators as they travel through their designated enumeration areas. The questionnaire may be in either paper or electronic form and various response modes may be used, such as mail-back, pick-up, telephone and Internet.

The various approaches involving administrative registers can range from all of the census data being produced from registers to only basic data being produced from registers, with the rest produced from complete enumerations and/or sample surveys, either ad hoc or pre-existing. At a minimum, however, methods 2.(a) to 2.(d) refer to an approach where at least the basic population and dwelling counts are produced from administrative registers.

Step 4

In the fourth step, an initial assessment was conducted on the applicability of each approach in the Canadian context, based on the questions listed in the Introduction. This assessment resulted in the recommendation to eliminate some of the options from further consideration for the 2016 Census.

Step 5

In the final step, recommended options to be retained for the 2016 Census were developed, as well as considerations for investigating and assessing approaches for future censuses (2021 and beyond).

Some important points should be kept in mind when reading the remainder of the report:

  • Each country has its own unique conditions for census-taking, including its legislative and administrative environment, its history with census-taking and government data collection generally, its specific information needs, and its financial resources. By classifying the wide variety of census-taking methods into a small number of basic methods, there is some risk that details that are important in a particular country context may be obscured. Nevertheless, basing the discussion of methods on the options listed above was found to be a useful approach.
  • It is as yet unknown exactly how the new 2011 Census and NHS approach will affect the possible options for conducting the 2016 Census. In particular, there is a scarcity of information or experience internationally on conducting a census where part or all of it is voluntary. The only known experience appears to be a 2003 test of a voluntary American Community Survey in the United States, and census questions on religion, which are optional in some countries.
  • This report addresses census methodology at a high level, and is not concerned with specific census technologies for data collection, data capture, data processing or dissemination. For example, it does not cover subjects such as questionnaire design, data collection by Internet or hand-held devices, data capture via optical character recognition, new edit and imputation methods, and record linkage methods. Instead, the census methodologies discussed in this report are concerned with the fundamental approach to producing data on the population and housing stock of Canada.
  • It is assumed in this report that the 2016 Census program will be national in scope and that the content that is required in Canada in 2016 is similar to that collected in the recent past by the census or, for 2011, by the combination of the 2011 Census and the NHS.
  • The UN, the UNECE and the CES have not yet developed their principles and recommendations for the 2020 round of censuses. While Statistics Canada is well represented in these bodies and continues to keep abreast of developments in other countries and international organizations, the international principles and recommendations referred to in this report largely pertain to the 2010 round of censuses.
Date modified: