Technical Guide on Demographic Estimates at Statistics Canada
Introduction

Demographic estimates are the cornerstone of statistical measurement for the population and are put to a wide variety of uses to gain a better understanding of the economic and social situation in Canada. Their importance is highlighted by the statutory requirements for the calculation of revenue transfers and cost-sharing programs between the various levels of government, and their role as population control totals adopted by the Labour Force Survey to derive primary economic indicators such as GDP per capita. Statistics Canada, the national statistical agency responsible for this, and other data, considers the timeliness, accuracy and reliability of demographic estimates to be of the utmost importance.

Every five years, Statistics Canada conducts a national census of population which collects a wide range of demographic data on the Canadian population. The census collects information on citizens, immigrants and non-permanent residents (NPRs) residing in Canada. While some countries have a system of continuous population registration, from which basic demographic data on the state and movement of the population for non-census years can be obtained, this is not the case in Canada.

With its five-year intervals, the census cannot meet the need for timely data required by Statistics Canada's various statistical programs and external data users. To fill this gap, Statistics Canada has developed the Demographic Estimates Program. Various methodological techniques use census data collected by Statistics Canada, along with administrative data provided by other government departments and organizations, to produce estimates of the Canadian population between censuses.^Note 

As such, demographic estimates are not to be mistaken for the Census counts. They represent an independent set of products of the Demographic Estimates Program derived from Census to establish a complete picture of the Canadian population and what factors are driving it to change over longer and shorter periods of time which the Census cannot capture. In addition to being produced more frequently, population estimates build upon raw Census counts with demographic adjustments and corrections for the Census net undercoverage, including that resulting from incompletely enumerated reserves and settlements. The page Differences between Statistics Canada's census counts and population estimates^Note  is available and contains more information on those differences.

In 1987, Statistics Canada published its first manual detailing the methods used for estimating population, Population Estimation Methods, Canada (Catalogue no. 91-528-X). The manual represented a direct response to a long-standing policy at Statistics Canada that requires the methods used to produce the agency's statistical information be open to public scrutiny. The manual was updated over time to document conceptual and methodological changes that have been made to the Demographic Estimates Program.

Data quality indicators and analysis of demographic estimates are important for users of these estimates and are helpful to Statistics Canada in fulfilling its role as the national statistical agency responsible for this data. Quality indicators are available inside this guide and within data tables released by Statistics Canada.

This technical guide replaces previous versions of the Population and Family Estimations Methods at Statistics Canada (Catalogue no. 91-528-X) manual. It also incorporates methods and concepts previously found in three other publications covering demographic estimates at Statistics Canada: Quarterly Demographic Estimates (Catalogue no. 91-002-X), Annual Demographic Estimates: Canada, Provinces and Territories (Catalogue no. 91-215-X), and Annual Demographic Estimates: Subprovincial Areas (Catalogue no. 91-214-X).

The guide intended to be a compendium of all methods and current procedures used by Statistics Canada to produce and release population and family estimates by combining and updating information from previous manual with that of the separate publications mentioned above.

Target population

Population counts from the Census form one of the fundamental building blocks necessary to produce population estimates. As such, the target population for demographic estimates is the same as that of the Census. Central to the definition of the target population are concepts of population universe - those that should be counted - and the usual place of residence – the definition of where an enumerated person should be allocated. The next sections contain description of those crucial concepts, based on excerpts from the Census Technical Report, Census of Population, 2021.^Note 

Population universe

The definition of target population specifies which persons should be included in the census, but not where these persons should be enumerated. The Canadian census uses the modified de jure method of enumeration, under which persons are to be enumerated at their usual place of residence (see next section), even if they are temporarily away on Census Day. Persons away from their usual place of residence and residing elsewhere in Canada must be enumerated at their usual place of residence and are considered present, but temporarily at the other location. Persons who have no usual place of residence are to be enumerated wherever they happen to be on Census Day.

The definition above is extended to demographic estimates produce at Statistics Canada, whereas the usual place of residence obtained from official administrative data sources is used to specify the location or characteristics of demographic events, regardless of temporary change in location.

The Census target population, and consequently the target population for demographic estimates at Statistics Canada, includes the following groups:

• Canadian citizens and landed immigrants (permanent residents) with a usual place of residence in Canada;
• Canadian citizens and landed immigrants (permanent residents) who are abroad, either on a military base or attached to a diplomatic mission;
• Canadian citizens and landed immigrants (permanent residents) at sea or in port aboard merchant vessels under Canadian registry or Canadian government vessels;
• non-permanent residents with a usual place of residence in Canada:
  • who are claiming refugee status (asylum seekers);
  • who hold a study permit (covering Census Day);
  • who hold a work permit (covering Census Day);
• and family members living with them.

The Census and demographic estimates population universe does not include foreign residents, but since 1991, it has included non-permanent residents.

Usual place of residence

Under the de jure enumeration method used in the Canadian census, the population is enumerated on a “usual place of residence” basis, that is, at the location where a person lives most of the time.^Note  It is used to identify the person as a member of a particular household and, potentially, family (depending on the composition of the household). Most individuals have only one residence, and it is easy to enumerate them at their usual place of residence. In some cases, it is difficult to determine one person’s usual place of residence. Special rules were developed to determine the usual place of residence on Census Day, the details of which can be found in the Census Technical Report, Census of Population, 2021 mentioned above.

In the case of demographic estimates, various administrative data obtained by Statistics Canada is used to create components of population change. With those, in conjunction with Census counts and several methodological adjustments, an estimated population can be derived for reference dates beyond the latest Census. The fundamental concept of demographic estimates rests on the availability of that additional administrative data which include, among others, Vital Statistics on births and deaths from provincial and territorial partners, fiscal information from the Canada Revenue Agency, and permanent and temporary immigration records from Immigration, Refugees and Citizenship Canada.

The individual demographic events contained within this administrative data provide the basis needed to estimate how population changes over time through migration between two regions, a family having a child, a foreign student coming to Canada, and more. The location of those events is significant as it enables to appropriately allocate changes in population to the correct province, territory, or subprovincial area. The most recent address information contained within those individual administrative records for a given reference date is used to represent the usual place of residence associated with each demographic event.

Gender and sex at birth

The demographic estimates now incorporate the concept of gender, which was first introduced in the 2021 Census of Population.^Note 

Demographic estimates are published by gender rather than by sex at birth, in accordance with the Treasury Board of Canada Secretariat's "Policy Direction to Modernize the Government of Canada's Sex and Gender Information Practices" (2018).^Note  This directive states that the gender variable is expected to be used by default in most social statistics programs at Statistics Canada.^Note 

Gender refers to an individual's personal and social identity as a man, woman or non-binary person (a person who is not exclusively a man or a woman). Gender includes the following concepts:

• gender identity, which refers to the gender that a person feels internally and individually;
• gender expression, which refers to the way a person presents their gender, regardless of their gender identity, through body language, aesthetic choices or accessories (e.g., clothes, hairstyle and makeup), which may have traditionally been associated with a specific gender.

A person's gender may differ from their sex at birth, and from what is indicated on their current identification or legal documents such as their birth certificate, passport or driver's license. A person's gender may change over time. Some people may not identify with a specific gender.

Demographic estimates are available for the gender category "Men+," which includes men (and/or boys), as well as some non-binary persons, and for the gender category "Women+," which includes women (and/or girls), as well as some non-binary persons.

For the sake of uniformity, all data products from the Demographic Estimates Program now adopt the term "gender" for all years and periods. Prior to 2021, the demographic estimates are based on the concept of sex at birth. Although gender and sex at birth are two different concepts, this change does not cause a significant break in the trend because the two concepts produce very similar distributions.

Population estimates by age and sex at birth continue to be available upon request. These estimates can be used where information on sex at birth is needed, for example, for measuring some demographic and health indicators. 

Differences between Statistics Canada's census counts and population estimates

The 2021 Census counted 36,991,981 people in Canada during the national enumeration with reference date May 11, 2021. This count is lower than the adjusted census population of 38,192,700 people calculated for the same reference date and which is used as a basis for population estimates. The difference between the two figures is not unexpected and is similar to that which was experienced for previous censuses. This note outlines why there are differences between census counts and population estimates. 

The objective of a census is to provide detailed information on the population at a single point in time. In this respect, one of its goals is to enumerate the entire population. Inevitably, however, some people are not counted, either because their household did not receive a census questionnaire (for example, if a structurally separated dwelling is not easily identifiable) or because they were not included in the questionnaire completed for the household (for example, the omission of a boarder or a lodger). Some people may also be missed because they have no usual residence and did not spend census night in any dwelling. In contrast, a small number of people may also be counted more than once (for example, students living away from home may have been enumerated by their parents and by themselves at their student address). 

To determine how many individuals were missed or counted more than once, Statistics Canada conducts postcensal coverage studies of a representative sample of individuals. Results of these studies in combination with the census counts are used to produce population estimates which take into account net undercoverage. 

For the 2021 Census, final coverage studies have been released on September 27, 2023. In turn, these have been used to revise and update the population estimates based on the 2021 Census results. Consequently, a series of revised population estimates for the period 2016 to 2023 has been released on September 27, 2023. 

One of the advantages of the census is to provide counts for small regions for which demographic estimates are not available or are less precise. On the other hand, population estimates provide more frequent measures of population counts for more aggregated levels of geography. In addition, estimates are used to measure the evolution of the population between censuses and provide explanations behind population growth.

Levels of estimates

There are two categories of population estimates produced by Statistics Canada: postcensal and intercensal. The first category, postcensal estimates, are produced by using data from the most recent census (adjusted for census net undercoverage (CNU), including incompletely enumerated reserves and settlements) and estimates of the components of population change since that last census. These components include births, deaths, immigrants, non-permanent residents, emigrants, returning emigrants, and interprovincial migrants – plus intraprovincial migrants for subprovincial demographic estimates.

Due to data timeliness and completeness, components mentioned above themselves can be updated over time as more complete data becomes available. This is represented by data being assigned a “level of estimates”, which differs slightly between components of population change and population estimates themselves. Each component will first be released as “preliminary” (noted “P”) level, with following releases moving to “updated” (noted “R”) or “final” (noted “D”) when new data is incorporated in following production.

Postcensal population estimates will be assigned a level based on that of the components used to derive it. Postcensal population estimates for any given reference date are always of level “preliminary postcensal” estimates (noted “PP”) when they are first released. Any change to the estimates resulting from one or more updated component of population change during a subsequent production would result in “updated postcensal” estimates (noted “PR”). Only when all individual components are considered “final” can the population be considered “final postcensal” estimates (noted “PD”).  

For the provinces and territories, preliminary estimates are typically available within three to four months after the reference date (seven to eight months for subprovincial estimates), while updated estimates are usually available within one year (one and a half year for subprovincial estimates). Final estimates are the most accurate postcensal estimates available and typically take two to three years to complete. The production of three types of estimates is the strategy that best satisfies the commitment of Statistics Canada to balance the timeliness and accuracy of data quality. 

Intercensal estimates are produced every five years and reconcile previous postcensal estimates with the latest census counts adjusted for census net undercoverage. The reconciliation with postcensal estimates leads intercensal estimates to need an additional component of population change, residual deviation. Those estimates are generated as soon as census population counts and census net undercoverage become available, a process which typically takes two years after census data collection to complete. Postcensal estimates are always assigned the level of “final intercensal” (noted “ID”) as they do not undergo further updates.

Disseminated estimation products

Demographic estimates are disseminated at six geographic levels: provinces and territories, census divisions, census metropolitan areas and census agglomerations, economic regions, health regions and peer groups, and census subdivisions. Demographic estimates at custom-defined subprovincial levels are possible and can be produced through special cost-recovery tabulations.

According to the level of geography and type of estimate, different demographic characteristics of the population (including age and gender) are produced. Table 1 shows the level of detail and the reference period for which demographic estimates are disseminated. 

Users of demographic estimates

At Statistics Canada, demographic information is used to calibrate sampling weights of many social and household surveys and is a fundamental part of the analytical framework of most statistical programs. Estimates of Canada's population have other wide-ranging applications in the areas of planning and program evaluation in both the public and private sectors. The calculation of revenue transfers under various federal statutory programs, for example, as well as cost-sharing agreements between federal, provincial, territorial and municipal governments are highly dependent on demographic data.

Statistics Canada is also under statutory obligation to provide the federal government with annual population figures as well as various economic indicators (e.g., gross domestic product) that have been certified by the Chief Statistician of Canada. These figures are used to determine the amounts payable under various federal–provincial fiscal arrangements such as the Equalization and Territorial Formula Financing (TFF), the Canada Health Transfer (CHT) and the Canada Social Transfer (CST), according to a per capita funding formula. The federal government distributes billions of dollars annually in federal transfers to provinces and territories using population estimates.^Note  Since 2011, the estimates have also been used in the provincial readjustment of federal electoral boundaries as described in the Fair Representation Act.^Note 

The CHT and CST are federal transfers that support specific policy areas, such as health care, postsecondary education, social assistance and social services, early childhood development, and childcare. 

The equalization and TFF programs provide unconditional transfers to the provinces and territories. Equalization allows less prosperous provincial governments to provide their residents with public services that are reasonably comparable to those in other provinces, at reasonably comparable levels of taxation. TFF provides territorial governments with funding to support public services to account for the higher cost of providing programs and services in the Canadian North. 

Population is a key variable of the distribution formula used by the federal government to determine its annual expenditures. As billions of dollars depend directly upon these allocation formulas, accurate and up-to-date estimates are extremely important. 

The Government of Canada relies on population data to make informed decisions on some of its most fundamental policies. For example, up-to-date information on the evolving demographic situation in Canada proves useful for Cabinet in its decisions relating to immigration policy. These data also contribute to increasing the cost-effectiveness of public policy expenditures. 

Provincial and municipal governments also use demographic estimates to plan social programs and establish cost-sharing agreements based on the per capita spending formula. Among the public service users of Statistics Canada's estimates are education and public health planners, and public administrators responsible for policing, criminal justice, municipal administration, and waste and environmental management. 

Demographic estimates are fundamental in the calculation of social and economic indicators, in particular birth rates, death rates, school enrolment rates, unemployment rates, and life expectancy. Population serves directly as the denominator in many of these indicators, with Statistics Canada directly providing population control totals used by the Labour Force Survey for its measures of economic indicators for example. Survey researchers, whether at Statistics Canada, in academia, or in private polling agencies, must use up-to-date figures in survey planning and in the calculation of sampling weights. 

The private sector also uses demographic estimates for business planning, marketing research and investments. The estimates also help companies properly segment and target their markets and make sound investment decisions. 

Overview of this guide

The Population estimation methods – Canada, provinces and territories chapter presents the overall methods used to produce population estimates for Canada, provinces and territories, and focuses on describing the various types of estimation - postcensal and intercensal estimates. The Population estimation methods – Subprovincial regions chapter covers various methods used to produce population estimates for subprovincial areas.

The individual chapters that follow include discussions on data sources, concepts and methodology for each of the components of population change. Each chapter describes how total estimates are produced, as well as estimates by age and gender, and subprovincial estimates for the component. This should give the reader a clearer picture of the limits of each method.

Topics covered include calculation of the base population, and the methods used for estimating births and deaths, the components of international migration (immigrants, non-permanent residents, emigrants and returning emigrants), interprovincial migration, and intraprovincial migration for subprovincial areas.

Later chapters describe the methods for estimating census families, economic entities and households, and the methods used to estimate and ensure data quality of demographic estimates over time. A glossary of the main terms used, and acknowledgements, are also available at the end of the document.