Estimates of the number of Census Families for Canada, Provinces and Territories

Detailed information for July 1, 2002

This estimates program provides annual estimates of Census families for Canada, Provinces and Territories.

Data release - March 4, 2003

• Description
• Data sources and methodology
• Data accuracy

Description

This estimates program provides annual estimates of Census families for Canada, Provinces and Territories.

This estimates program is used in the calculation of demographic, social and economic indicators. These estimates are used for planning, program evaluation and base population for various surveys and studies.

Subjects

• Population and demography
• Population estimates and projections

Data sources and methodology

Target population

For the purpose of generating estimates, the definition of family is, with one exception, similar to what is used for the Census. As defined up to and including the 2006 Census, a census family can be any of the following:

- a married couple (with or without children of either or both spouses);
- a couple living common-law (with or without children of either or both partners) ;
- a lone parent of any marital status, with at least one child living in the same dwelling.

The term children refers to blood, step- or adopted sons and daughters (regardless of age or marital status) who are living in the same dwelling as their parent(s), as well as grandchildren in households where there are no parents present. Sons and daughters who are living with their spouse or common-law partner, or with one or more of their own children, are not considered to be members of the census family of their parent(s), even if they are living in the same dwelling. In addition, the sons or daughters who do not live in the same dwelling as their parent(s) are not considered members of the census family of their parent(s). When sons or daughters study or have a summer job elsewhere but return to live with their parent(s) during the year, these sons and daughters are considered members of the census family of their parent(s).

The exception resides in how same-sex couples are recognized. While in census, same sex common-law partners are creating a census family, this is not the case for the estimates program.

Instrument design

This methodology does not apply.

Sampling

This methodology does not apply.

Data sources

Data are extracted from administrative files and derived from other Statistics Canada surveys and/or other sources.

Total Number of Families - The change in the number of families between any two points in time (t,t+i) is determined by events which result in family formation or dissolution. These events, classified by their impact on number of families, are as follows: information on marriages, divorces, deaths, immigration, total emigration, non-permanent residents and interprovincial migration. The important family formation/dissolution factors are either readily available, or derived indirectly from administrative data sources. These sources, however, lack information on common-law unions, defacto separations, marriage annulments, and on children leaving home - making a complete accounting of factors contributing to the formation (or dissolution) of a family almost impossible. Fortunately however, the events for which little or no statistical information is available are infrequent, and their anticipated impact is small. The breakdown of the total number of families by sex and age of parent, and further by family size and children's age is accomplished on the basis of data from the most recent census of population.

Error detection

This methodology type does not apply to this statistical program.

Imputation

This methodology does not apply.

Estimation

The method to generate postcensal estimates of the total number of families by province and territory and by characteristics for provinces only consists of two steps: 1) Total provincial/territorial number of families is obtained by the component method. The events responsible for a formation and dissolution of families which occurred between July of census year Y and the reference date of the estimates are added to or subtracted from the year Y Census figures adjusted for net undercoverage. The following three types of events are accounted for: events contributing to formation of families resulting from marriages: events resulting in dissolution of families due to divorce, death of married persons, death of widowed persons and death of divorced persons; events triggering net change in the number of families (net interprovincial and international migration of families, net change in non-permanent resident families, returning Canadian families and net change in common-law unions); 2) Total number of families estimated in the previous step is distributed according to family type, sex and age of parents, family size and children's age by extrapolating changes in these distributions as observed between the two previous censuses.

Quality evaluation

One method of evaluating family estimates is to compare the estimates for the census year to the census counts. There is a considerable variation in the error of closure for families by province/territory. There is also variation when comparing the error of closure for the two different intercensal periods. For instance, in 1996, the overall error of closure was almost half of that for 1991. Reduced error for 1996 estimates was most noticeable for Quebec, Manitoba and Saskatchewan. However, the error of closure was considerably higher (more than double) in 1996 for Nova Scotia and the Northwest Territories. A portion of this difference may be explained by the fact that the 1991 estimates (both Demography Division's and the censal estimates) do not account for an adjustment for net census undercoverage of families and could be affected by the differentiated census coverage of 1986 and 1991. The 1996 estimates do include this adjustment, therefore they are affected only by errors in the estimates of the completeness of families' enumeration in both 1991 and 1996.

For more detailed information on the quality evaluation of the estimates of Census families, see Chapter 10 of Population and Family Estimation Methods at Statistics Canada, Demography Division, Catalogue 91-528-XIE.

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.

In order to prevent any data disclosure, confidentiality analysis is done using the Statistics Canada Generalized Disclosure Control System (G-Confid). G-Confid is used for primary suppression (direct disclosure) as well as for secondary suppression (residual disclosure). Direct disclosure occurs when the value in a tabulation cell is composed of or dominated by few enterprises while residual disclosure occurs when confidential information can be derived indirectly by piecing together information from different sources or data series.

Revisions and seasonal adjustment

Data are revised once a year and after each census, postcensal estimates are revised to produce the intercensal estimates.

Demographic estimates are revised using birth, death and interprovincial and international migration statistics when they become available. Revisions may result in notable changes for certain components, particularly for interprovincial migration.

Interprovincial migration data are derived from two sources. Preliminary migration estimates are based on changes of addresses recorded by the Canada child benefit program from the Canada Revenue Agency, and are available shortly after the reference month. Final interprovincial migration estimates are based on addresses supplied on personal income tax returns, and are available a year after the reference year.

Data accuracy

Four sources of error and bias contribute to the error of closure: (1) common-law unions; (2) the correction for census net undercoverage; (3) the universes covered; and (4) unrecorded events. Common-law unions - While common-law unions are counted as "marriages" in the censuses, they are not included in Vital Statistics. Family universes - The family universes for the census enumeration and Vital Statistics are not identical, due to differential reporting of families in collective households. These families are not included in census family tabulations. However, the events occurring to members of families or non-family persons in collective households are registered by Vital Statistics, resulting in a discrepancy between families as identified by the census and the events attributed to these families. The magnitude of error associated with this discrepancy is unknown. Unrecorded events - unrecorded events, such as: the death or departure from home of the only child of a lone-parent family; separations, annulments of marriages; and births of children to non-family persons, are also possible sources of error. Some have a positive, others a negative effect on family formation. The net effect, however, is probably insignificant.