Skip to content | Skip to institutional links

Common menu bar links

Institutional links

Census

Page content follows

View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

Mobility and Migration Reference Guide, 2006 Census

Catalogue no. 97-556-GWE2006003

Definitions and explanations of variable concept(s)

Introduction to the population universe

There are two subuniverses for Mobility and migration: Mobility status 1 year ago, which is based on the answers to the question on the place of residence 1 year prior to the census date (Question 23) and Mobility status 5 years ago, which is based on the answers to the question on the place of residence 5 years prior to the census date (Question 24). These subuniverses exclude institutional residents, Canadians who are in households abroad such as diplomats and soldiers stationed abroad. Further excluded are persons under 1 year of age for the Mobility status 1 year ago subuniverse and persons under 5 years of age for the Mobility status 5 years ago subuniverse.

There are 12 variables each in the Mobility status 1 year ago and in the Mobility status 5 years ago subuniverses.

Questions

Mobility and migration data were obtained from Questions 23 for 1 year ago and 24 for 5 years ago on the census long (2B and 2D) questionnaires. The 2B questionnaires were used to enumerate a 20% sample of all private households in Canada, while the 2D questionnaires were used to enumerate private households on Indian reserves and in remote areas. The mobility questions are Questions 23 and 24 on the 2006 Census 2B questionnaire.

Classification

Mobility status – Place of residence 1 year ago

Refers to the relationship between a person's usual place of residence on Census Day and his or her usual place of residence one year earlier. A person is classified as a non-mover if no difference exists. Otherwise, a person is classified as a mover and this categorization is called Mobility status (1 year ago). Within the movers category, a further distinction is made between non-migrants and migrants; this categorization is called migration status.

Figure 11 Relationship between the category of mobility status and the 2006 Census question on place of residence 1 year ago

Non-movers are persons who, on Census Day, were living at the same address as the one at which they resided one year earlier.

Movers are persons who, on Census Day, were living at a different address than the one at which they resided one year earlier.

Non-migrants are movers who, on Census Day, were living at a different address, but in the same census subdivision (CSD) as the one they lived in one year earlier.

Migrants are movers who, on Census Day, were residing in a different CSD one year earlier (internal migrants) or who were living outside Canada five years earlier (external migrants).

Mobility status – Place of residence 5 years ago

Refers to the relationship between a person's usual place of residence on Census Day and his or her usual place of residence five years earlier. Classifications of respondents are the same as for Place of residence one year ago described above.

Figure 12 Relationship between the category of mobility status and the 2006 Census question on place of residence 5 years ago

The place of residence within Canada is coded to the Standard Geographical Classification (SGC) 2006, a 7-digit code for each census subdivision.

Data

Mobility and migration

Data Quality

Data quality verification in place for the 2006 Census

Processing

Processing of the data described in this guide includes the coding of the responses and the edit and imputation of invalid responses and non-responses.

Coding of Mobility

The mobility questions contain four fields requiring automated coding:

Place of residence 1 year ago within Canada
Place of residence 1 year ago outside Canada
Place of residence 5 years ago within Canada
Place of residence 5 years ago outside Canada

A reference file for place name is used for coding place of residence within Canada. For the first time in the 2006 Census, postal code responses were added for coding the place of residence. If both postal code and place name write-in responses exist, both were used to code with the postal code taken as final.

The place of residence within Canada is coded to the Standard Geographical Classification (SGC) 2006.

A reference file for name of countries is used for coding place of residence outside Canada, using three-digit codes assigned to each country.

Non-response and invalid responses

Non-response means that there were no marks in any of the boxes and no entries in any of the write-in areas.

The term 'invalid response,' refers to one or more of the following:

a response in which more than one box was marked
an inconsistent response, in which the marked box conflicts with the write-in response
a census subdivision or country that cannot be coded.

For the question on Place of residence 1 year ago in the 2006 Census, the non-response rate was 2.3% and the invalid response rate was 1.5%, an increase from 1.3% and 0.9%, respectively, in 2001 Census.

For the question on Place of residence 5 years ago, the non-response rate was 2.3% in 2006, an increase from 1.6% in 2001. Similarly, the invalid response rate increased from 1.3% in 2001 to 2.4% in 2006.

Non-response and invalid response rates were lower when an Internet version of questionnaires was used, compared with paper questionnaires. The non-response rate for Place of residence 1 year ago was 0.6% for the Internet version, as opposed to 1.8% for the paper version. The invalid response rate was also lower at 0.0% for the Internet version compared with 1.5% for the paper version.

The results were similar for the Place of residence 5 years ago question: 0.7% non-response rate for the Internet version versus 1.8% for the paper version; 0.0% invalid response rate for the Internet version versus 2.6% for the paper version.

Imputation for non-response and inconsistent responses

During automated data processing, the quality of the data on mobility is maximized by checking the logical consistency of the different characteristics. Then, any invalid, inconsistent or missing responses are replaced with acceptable values through a process called 'imputation.' Two types of imputation are applied on mobility data, namely deterministic imputation and donor imputation. Deterministic imputation is the process by which a unique value is assigned to a missing or inconsistent response through relationships among characteristics. Donor imputation is performed by identifying individuals in the same geographical area that have similar, but complete and consistent characteristics and then copying the values of randomly selected individuals to fill in the missing or erroneous data among the 'failed edit' individuals.

The Canadian Census Edit and Imputation System (CANCEIS) was used for the pre-derivation and post-derivation modules where imputations of mark-in and write-in responses in cases of inconsistent responses are performed. Edit and imputation procedures resolve non-responses, invalid responses and assigns mobility values for persons under 15 years of age.

For the questions on Place of residence 1 and 5 years ago in the 2006 Census, the percentages of deterministic imputation and CANCEIS imputation (i.e., donor imputation) are not much different from those of the 2001 Census.

For Place of residence 1 year ago for all age groups, the CANCEIS imputation rate was 17.9% both in the 2006 Census and in the 2001 Census, while deterministic imputation was slightly lower in the 2006 Census at 1.2% than in the 2001 Census at 1.7%.

For Place of residence 5 years ago for all age groups, the CANCEIS imputation rates were slightly lower at 14.9% in the 2006 Census than at 16.3% in the 2001 Census, and the deterministic imputation rates were also slightly lower in the 2006 Census at 3.5% than in the 2001 Census at 3.9%.

The reason for higher rates of CANCEIS imputation than deterministic imputation is mainly due to the fact that CANCEIS imputation was performed in assigning mobility information to the children under 15 years of age who are not required to answer to the census question.

This can be seen at the very high rate of CANCEIS imputation for Place of residence 1 year ago for the persons under 15 years of age. In 2006, this rate was 95.0%, and in 2001, it was 84.8%. In comparison, the CANCEIS imputation rates were only 2.8% in 2006 and 2.1% in 2001 for persons 15 years of age and over. In 2006, for Place of residence 5 years ago, this rate was 92.1% and in 2001 it was 96.6%. In comparison with the children under 15 years of age, the CANCEIS imputation rates were only 3.7% in 2006 and 3.2% in 2001 for the persons 15 years of age and over.

More information on edit and imputation is available in the 2001 Census Technical Report on Mobility and Migration.

Data comparability

The mobility question on Place of residence 1 year ago was first introduced in the 1991 Census. At this time, only the province of residence 1 year ago was asked. From the 1996 Census on, the geographic levels covered were identical to the question of Place of residence 5 years ago, in that the census subdivision (CSD) has been used as the migration-defining boundary unit.

The mobility question on Place of residence 5 years ago has not differed significantly from the five-year questions of previous censuses. Therefore, the five-year mobility data are generally comparable from 1961 on. The question has been based on a five-year reference interval and the census subdivision (CSD) has been used as the migration-defining boundary unit.

The number of migrants, that is, persons who crossed the census subdivision boundary, is affected by the number of census subdivisions which is different from one census to another. Due to amalgamation of municipalities (CSDs ), the number of CSDs has been reduced from 5,984 in 1996 through 5,600 in 2001 to 5,418 in 2006.

Comparability of mobility data among the various censuses is described in greater detail in the 2006 Census Dictionary, Appendix I.

Comparison with other data sources

As with every census, the quality of the 2006 Census mobility and migration information released was evaluated internally prior to publication. The data were compared, to the extent possible, with alternative data sources. The main source for comparison was two folds: the migration estimates from income tax files for internal migration data and administrative files from Citizenship and Immigration Canada for external migration data. For the comparison of interprovincial migration data, the final estimates of migration based on income tax files were used for both 1- and 5-year data. For census metropolitan areas, the last one year data (2005-2006 period) were preliminary estimates based on child tax benefit data. For the comparison of external migration, the landed immigration data and the student and working visa authorization files obtained from Citizenship and Immigration Canada were used.