First report on the Census Program Transformation Project:
Researching a new approach to census-taking

What's new in census research?

The 2016 Census Strategy Project was conducted from 2010 to 2012. It examined three approaches to census-taking used by countries around the world: the traditional census, the use of existing administrative registers and the use of continuous measurement.

After examining the benefits and challenges of these approaches in the Canadian context, the Final report on 2016 Census Options concluded that the traditional census approach was the only viable methodology for the 2016 Census Program, notably because of time constraints. The report recommended continued long-term research on alternative methodologies for the Census Program, in particular the increased use of administrative data.

Since 2013, Statistics Canada has continued its research and built prototypes of a statistical population database—known as the Canadian Statistical Demographic Database—in an effort to reproduce the 2011 Census head count. This database was built using administrative data obtained primarily from the Canada Revenue Agency and Immigration, Refugees and Citizenship Canada, as well as provincial and territorial vital statistics under the authority of the Statistics Act. It contained basic demographic information such as sex, date of birth and an address for each person in Canada with the goal of being used for statistical purposes only. The counts from the database were compared with the 2011 Census counts and the 2011 Censal Demographic Estimates. They compared well with them at the national, provincial and territorial levels, as well as by age and sex. However, the research found that more work was needed at lower levels of geography to achieve the same quality as the census and that more analysis of sub-populations was required.

For 2016, Statistics Canada built other versions of the Canadian Statistical Demographic Database, enhancing it with additional administrative data to ensure all Canadians are placed in the right location. A thorough analysis is being conducted to assess the efficiency and quality of the 2016 prototypes.

What is the Census Program Transformation Project?

The long-term research recommended in the Final report on 2016 Census Options culminated into the Census Program Transformation Project (CPTP), which was launched in June 2016. The CPTP continues to explore how the use of administrative data can support the production of the census population and dwelling counts, but it does so in a more holistic way than previous research.

The CPTP is investigating the development of a new combined census methodology to produce population and dwelling counts, while ensuring that Canadians continue to receive the high-quality information they expect from the Census Program. This new methodology would leverage available administrative data rather than collecting the information directly from Canadians. The CPTP's overall objective is to improve the efficiency of the census by building robust and interconnected statistical registers on buildings and population using a variety of administrative data sources.

Research into this new approach will assess how current methodology for the short- and long-form census would have to be adjusted to complement information not available in statistical registers. It will also assess options for measuring the coverage of population counts produced from a collection methodology that combines traditional census and administrative data from statistical registers.

Message to census stakeholders and Canadians

The use of statistical registers should reduce Census Program costs as well as the response burden placed on Canadians while maintaining overall quality and relevance. Statistics Canada will also be able to reuse the statistical registers to improve the efficiency, relevance and responsiveness of its social statistics programs.

Population counts and associated coverage errors are at the heart of Statistics Canada's Demographic Estimates Program. These estimates are used as part of the funding formula to determine the distribution of major federal transfer payments to the provinces and territories under the Federal-Provincial Fiscal Arrangements Act. The CPTP will ensure the high quality of demographic estimates is maintained with a combined census model.

As research progresses, Statistics Canada will continue to consult and engage with census data users, stakeholders and Canadians. The agency will provide ongoing updates on its research plans and progress, and will seek feedback during key phases of the research.

Canadians can rest assured that their privacy will continue to be protected; Statistics Canada treats administrative data with the same degree of scrutiny for confidentiality and security as information collected through its surveys. The research performed by CPTP falls within the scope of Statistics Canada's privacy impact assessment, which concluded that risks associated with the collection of personal information under the authority of the Statistics Act—including administrative data—are properly mitigated to protect the privacy of individuals.

Next steps

In 2017 and 2018, research efforts will focus on:

• developing the statistical registers
• establishing options to best combine information from statistical registers and census questionnaires to produce population and dwelling counts and estimates of other census demographic characteristics and to continue as such to respond to priority data needs; and
• determining options to measure coverage of the population counts under this new model.

The first test and assessment of the new combined census model will take place in 2019. If this assessment proves promising, portions of the model could be partially implemented in 2021 to support non-response follow-up efforts as part of the collection of short-form census questionnaires. The combined model will be further tested in 2021 to determine whether—and to what extent—it can replace the use of short-form census questionnaires in 2026 and future censuses.

More on statistical registers

More on a combined census model