Annual Migration Estimates by Census Division/Census Metropolitan Area

Detailed information for July 1, 2005 to June 30, 2006

Status:

Inactive

Frequency:

Annual

Record number:

4101

This statistical activity is conducted to produce annual migration estimates between census divisions, census metropolitan areas, provinces/territories, and movement to and from Canada, by age group and sex of migrants.

Data release - September 27, 2007

Description

The annual inter-provincial migration data are a legislated requirement to provide population estimates (The Federal-Provincial Fiscal Arrangements Act). This statistical activity is conducted to produce annual migration estimates between census divisions, census metropolitan areas, provinces/territories, and movement to and from Canada, by age group and sex of migrants. The data are provided to Demography Division for input to their population estimates. The data are used by provincial governments, demographers, statisticians, and business.

Reference period: Calendar year "y" for income, end of calendar year "y" for age, point in time (usually April of calendar year "y+1") for address information.

Collection period: The T1 files for income year "y" are received from Canada Revenue Agency (CRA) in January of year "y+2".

Subjects

  • Mobility and migration
  • Population and demography

Data sources and methodology

Target population

The statistical activity covers the migrant population between census divisions, census metropolitan areas, provinces and territories, and to and from Canada.

Instrument design

This methodology does not apply.

Sampling

This methodology does not apply.

Data sources

Data are extracted from administrative files.

The migration estimates are derived from a comparison of addresses from individual income tax returns for two consecutive years. The taxation records are obtained from Canada Revenue Agency (CRA). The period of reference for address information extends primarily from April of one year to April of the following year.

Error detection

During processing, the postal codes on records are verified for authenticity by comparison with a file containing authentic active codes. Those failing this check are then coded from the address. If that fails, the remaining records are coded from the municipality. Of the 24 million records processed in 2005, 99.9% contained legitimate active postal codes or can be identified as foreign. The remaining 23,092 records were assigned postal codes from their address or municipality.

Spurious migration is identified throughout data processing by machine, with elimination identified through manual examination and subsequent machine elimination.

Imputation

Because the source file has limited direct information on the number and characteristics of non-filing individuals, this information must be obtained. The current system uses the estimation of taxfilers' dependants from the T1 family file (T1FF). The family system creates families by linking filing family members together and estimates non-filing members from information on the taxfilers' returns, based on marital status, deductions and information for tax credits or from an historical file of children. For example, the family system imputes a non-filing spouse wherever a filer has declared him/herself married but was not linked with a filing spouse.

Between 1989 and 1992, information about children was derived directly from the tax file. Starting with the 1993 data, a combination of files was used to identify non-filing children: the Canada Child Tax Benefit file, the provincial births files and the T1FF of the previous year.

Approximately 70% of the Canadian population files a tax return. A completed T1FF accounts for approximately 96%, the difference being non-filers identified from filers' information.

Estimation

The production of the estimates involves the following main steps:

Geocoding of tax records -- The geographic coding of census divisions (CDs) and census metropolitan areas (CMAs) is done primarily on the basis of the postal code, which is part of the mailing address. In some cases, other pieces of information are used in place of a missing postal code. Essentially all records are assigned a CD code (when one includes a code for outside Canada) and a CMA/non-CMA identifier.

Identification of the number, age group and gender of migrants -- The migrants are identified by comparing current and previous CDs or CMAs of residence. The estimates of international migration are prorated to agree with provincial estimates provided by Demography Division of Statistics Canada.

Coverage Correction -- Each of the main source files used covers approximately 96% of the total population. The combined file which contains those people (and their non-filing dependents) who filed in two consecutive years contains approximately 87% of the population. The final step in the estimation process is an adjustment for coverage, done by age and gender at the CD and CMA level. Population estimates for July 1 of the first year of the migration period, by CD and CMA, are used to create coverage ratios. In a few cases where coverage is abnormally high or low, provincial adjustment ratios are used in place of the CD/CMA ratio. Beginning with 2001-2002 migration data, high and low coverage was identified with a new methodology and a Canadian adjustment ratio was used in place of the CD/CMA ratio.

Quality evaluation

The migration estimates are evaluated by trend analysis and are more closely scrutinized after each Census.

Disclosure control

Statistics Canada is prohibited by law from releasing any data which would divulge information obtained under the Statistics Act that relates to any identifiable person, business or organization without the prior knowledge or the consent in writing of that person, business or organization. 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.

The migration estimates themselves are modelled, with the tax file results weighted to represent the whole population. As such, the estimates are not absolute results from the tax file itself. When income information is provided, all counts are examined for minimum cell size of 15 and suppressed if they do not meet the requirement. The cells are rounded to the nearest 10. Medians are provided only when there are at least 30 people.

Revisions and seasonal adjustment

The counts of migrants obtained in the process are weighted according to the population estimates of July 1st for the beginning of the migration period. These weights are calculated by census division (or CMA), by sex and by five age groups. The international components are benchmarked to the provincial/territorial immigration/emigration data available at the time of production.

One revision is made to the data, that for the international components. Once the final immigration/emigration information is available, the international components are re-benchmarked to these figures - by province/territory, age group and sex.

Data accuracy

Overall, the estimates of migration are of good quality. In a simulation exercise using 1986 Census population as the base population, migration estimates have been used to produce population estimates for 1991 which were then compared to the 1991 Census counts. The average absolute difference was 2.3%. In 12 of 182 cases (6%) the deviation exceeded 5% and in 3 cases, the deviation exceeded 10% (this did not include Quebec census divisions). These deviations are smaller than those obtained from other estimation methodologies and indicate the quality of the migration estimates.

Documentation

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: