Preliminary Estimates of Population for Census Divisions and Census Metropolitan Areas (Regression method)

Detailed information for July 1, 1997





Record number:


This estimates program is used in the calculation of demographic, social and economic indicators (vital rates, unemployment rates, school enrolment rates, etc.) in which the population, or a part thereof, serves as the denominator.

Data release - -


This estimates program is used in the calculation of demographic, social and economic indicators (vital rates, unemployment rates, school enrolment rates, etc.) in which the population, or a part thereof, serves as the denominator. These estimates are also used in calculation of weights for use in Statistics Canada's Surveys. In addition, the data helps in the preparation of population projections by Statistics Canada, where estimates of population are used as the base population.

Reference period: July 1


  • Population and demography
  • Population estimates and projections

Data sources and methodology


Using the most recent estimates obtained by using the component method as a base, the regression method is used to estimate the change in population for census divisions (CDs) up to 1997. The regression method, employing either the ratio-correlation or the difference-correlation technique, is used to estimate any event (dependent variable) using a set of symptomatic indicators as predictors (independent variables). The regression models for each province utilize the best available symptomatic indicators of population change, namely: the number of child tax benefit recipients aged 1-14 for most provinces and the population registered in the provincial health insurance programs in Saskatchewan and Alberta. The regression estimates of population for CDs in British Columbia are produced by the Ministry of Finance and Corporate Relations. Different regression models maximize the accuracy of population estimates by taking advantage of administrative files containing data on specific local areas. Population estimates for census metropolitan areas (CMAs) are obtained from CD estimates using conversion factors derived from census counts of census subdivisions , geographical units common to both CDs and CMAs.

Preliminary postcensal estimates of CDs and CMAs from 1998 to the present are produced using the Rate of Growth Method.

The rates are based on the average annual growth estimated over the last two-year period. The CD and CMA numbers are controlled to the July 1st provincial and territorial estimates.

The preliminary estimates of population for CDs and CMAs in Quebec and British Columbia are prepared by their respective statistical agencies. Estimates from "l'Institut de la statistique du Québec" are prepared using administrative files containing data on the number of persons registered in "La Régie de l'assurance maladie". Those of British Columbia are prepared by BC STATS based on a regression model using residential and electrical meters and Old Age Security data as symptomatic indicators. These estimates are controlled to Statistics Canada's provincial estimates.

The distribution by age and sex for CDs and CMAs are prepared by aging estimates of the previous year, assuming a natural increase equivalent to that of the previous year, and zero net growth for all other components. The age and sex data thus obtained are controlled to the corresponding CD and CMA totals as well as to the provincial and territorial estimates by age and sex.

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects that could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.

Data accuracy

The estimates of total population have a certain margin of error, the magnitude of which may vary from one area to another. The average absolute error for all CDs in Canada for the period 1991-1996 is 1.59%. The errors can originate from the adjustment of the census counts for net census undercoverage; the final estimates for previous years based on the component method; the symptomatic indicators used in the provincial regression models; and for CMAs, the CD/CMA conversion process. Unfortunately there is no direct way to measure the error in postcensal population estimates of census divisions and census metropolitan areas.


Report a problem on this page

Is something not working? Is there information outdated? Can't find what you're looking for?

Please contact us and let us know how we can help you.

Privacy notice

Date modified: