Technical note: Putting 2016 Census results in context

Release date: October 18, 2017Updated on: November 22, 2017

The census provides a detailed statistical portrait of Canadians every five years. From one census to the next, there are small changes in processes that can affect the comparability of data over time, such as changes to questionnaire definitions and design, respondent instructions, sampling, and data collection and processing. Changes also include using data from new sources to enhance census data quality and reduce response burden for Canadians.

Each census, Statistics Canada provides technical information so that data users can better understand the latest census results. Here are some factors to keep in mind for the release of 2016 Census long-form information on immigration and ethnocultural diversity, housing, Aboriginal peoples, education, labour, journey to work, language of work, and mobility and migration.

Reinstatement of the mandatory long-form questionnaire

The methodology used for the 2016 Census differs from the one used for the voluntary 2011 National Household Survey, which may affect the comparability of the 2016 results to 2011. The collection response rate for the 2016 Census long form was 97.8%, the best ever recorded. This response rate enables the provision of high-quality information for virtually all communities in Canada.

Immigration

Questions on citizenship, immigration and place of birth have undergone little change over time, but factors such as changes in questionnaire instructions and format, as well as applicable legislation should be taken into account when comparing the data to previous censuses.

Two new immigration variables were added for 2016—the admission category and applicant type—to provide information on the conditions under which immigrants landed in Canada. These new variables are the result of a record linkage with administrative immigration records for immigrants admitted since 1980. This new information will allow users to understand the socio-economic outcomes of immigrants according to their admission category (e.g., refugee, economic immigrant).

Ethnic origin

On the census questionnaire, Statistics Canada often adds examples to questions to illustrate the type of information that respondents should provide. In the case of the ethnic origin question, Statistics Canada updated the list of ethnic origin examples in 2016 to reflect the single responses reported most frequently in 2011, based on a long-established methodology for choosing examples for this question. The categories Iranian and Mexican were added to the list of examples presented on the 2016 questionnaire, while Jewish and Salvadorean were removed based on this methodology.

For the first time, in 2016, Statistics Canada is publishing data for the following 15 ethnic origins in order to provide more detailed information: Arawak, Bavarian, Bhutanese, Catalan, Corsican, Djiboutian, Edo, Ewe, Guadeloupian, Hazara, Karen, Kyrgyz, Malinké, Turkmen and Wolof. In past censuses, these ethnic origins had been grouped with other ethnic origins.

Visible minorities

The examples of population groups used in the questionnaire to derive the visible minority population were updated. For example, Thai replaced Malaysian as an example used for the Southeast Asian category in the 2016 Census.

Aboriginal peoples

As in previous censuses, Statistics Canada worked with First Nations, Métis and Inuit communities to encourage participation in the 2016 Census. Fourteen reserves or settlements were incompletely enumerated in the census, the lowest number since 1986.

Over time, legislative changes have affected the reporting of concepts such as Aboriginal identity or Registered or Treaty Indian status, including Bill C-31 in 1985 and Bill C-3 in 2011. For these reasons and others, some people report their Aboriginal identity or ancestry differently from one census to another.

Housing

Users should be aware of the following questionnaire changes for the housing questions when comparing housing results over time.

The wording of the 2016 Census question on the number of rooms/bedrooms in dwellings was slightly modified.

In 2011 and 2016, the questionnaires were modified to ask all private households to indicate whether their dwelling is part of a condominium project. Prior to 2011, only owner-occupied households had been asked this question. As well, questions were included to identify tenant households living in subsidized dwellings, whereas the 2006 Census had no questions on housing subsidies.

Income

For the 2016 Census, Statistics Canada gathered income information solely from administrative data sources for the first time. The use of administrative data not only reduced respondent burden, but also increased the quality and quantity of income data available.

In September 2017, the agency released income information for 2015, along with information on demographic, family and household characteristics obtained from the short-form census questionnaire. In upcoming census releases, Statistics Canada will publish income estimates for variables obtained from the long-form census questionnaire. These income estimates may differ from those published in September, because of sampling effects, as well as higher rates of imputation among specific subpopulations.

Education

While the educational concepts in the 2016 Census are the same as in past censuses, changes were made to the wording and presentation of the education questions in 2016 to improve the accuracy of reporting and/or reduce respondent burden.

In 2006 and 2011, information on postsecondary qualifications was collected in three separate questions on completed trades, college and university qualifications. In 2016, these responses were presented as parts a), b) and c) of a single question (Question 26). In addition, on the electronic version of the questionnaire, detailed response categories for these questions were presented only if the respondent selected "Yes" in response to the initial question.

Other changes for postsecondary qualifications included the addition of instructions for all three parts of Question 26 to emphasize that only completed credentials were to be reported. Question 26 also included examples of different university qualifications (e.g., B.A., B.Sc., B.Ed., and LL.B. for "bachelor's degree"; M.A., M.Sc., M.Ed. for "master's degree").

In addition, the electronic questionnaire included dynamic text in the major field of study (Question 27) and location of study (Question 28) questions that referred back to the highest educational qualification. For example, if the highest qualification reported was a bachelor's degree, Question 27 would ask the "major field of study of bachelor's degree," rather than using the default text "major field of study of the highest certificate, diploma or degree."

Major field of study information from the Census will be disseminated using various aggregations of the Classification of Instructional Programs (CIP) Canada 2016, including the newly updated STEM and BHASE groupings variant.

The STEM fields of study include science, technology, engineering and mathematics. The new variant includes a breakdown of the BHASE (non-STEM) grouping into sub-groupings which encompass business, humanities, health, arts, social science, education, legal studies as well as a category for trades, services, natural resources and conservation fields of study. The term 'non-STEM' in the previous classification variant has been replaced by the term 'BHASE.'

Please see the Statistics Canada website for additional details.

Labour

For both 2011 and 2016, the most detailed labour data by industry are available at the four-digit North American Industrial Classification System (NAICS) code level. However, in 2016, data were classified according to NAICS 2012, whereas in 2011 data were classified according to NAICS 2007.

The following NAICS codes were affected by a classification update or a change in coding strategy:

A complete concordance table for 2007 and 2012 NAICS classifications is available online.

For both 2011 and 2016, the most detailed labour data by occupation are available at the four-digit National Occupation Classification (NOC) code level. Data from 2016 were classified according to NOC 2016, whereas 2011 data were classified according to NOC 2011. The updates made to the classifications between 2011 and 2016 were minor.

However, between 2011 and 2016, Statistics Canada updated some of its data coding methods, and this may affect the comparability of the 2011 and 2016 data for the following NOC codes:

Journey to work

In the 2016 Census, Statistics Canada improved geographic coding for places of work in rural census subdivisions (CSD). This improvement means that data on distance from home to work are more precise for many rural CSDs than in past censuses.

Previously, places of work in rural CSDs were coded using a representative point in the largest block within the CSD. Although this practice continued for manual and some automated coding, part of the automated system was adjusted in 2016 to record places of work more precisely in a CSD blockface or block.

Language of work

The 2016 Census language data for mother tongue, home language and knowledge of official languages were originally disseminated on August 2, 2017. After the dissemination, Statistics Canada corrected anomalies in the original results, primarily for selected regions in Quebec. The language data were then re-released. A description of how the error was identified and the nature of the error is available in an Update of the 2016 Census language data published on the Statistics Canada website.

As a result of the corrections to these variables, data on the knowledge of non-official languages and on the language of work were reprocessed to ensure that any changes made to the mother tongue and home language variables were reflected in the data on knowledge of non-official languages and on language of work. For more information, refer to Appendix 1.8 of the Guide to the Census of Population, 2016, Catalogue no. 98-304-X on the Evaluation of the impact of updating 2016 Census language data.

Mobility and migration

The mobility questions are unchanged from 2011.

However, in 2016, mobility and migration data were collected from Questions 22 and 23. In 2011, they were collected from Questions 23 and 24.

Additional technical documentation

For more information on census concepts, definitions and methodology, consult the Release and Concepts Overview of Aboriginal peoples, housing, immigration and ethnocultural diversity, education, labour, journey to work, language of work and mobility and migration.

In addition, Statistics Canada offers reference guides to help users understand, interpret and use census data. Reference guides on the following topics will be released on November 29, 2017:

Other reference guides are available on Statistics Canada's website:

Date modified: