Sampling and Weighting Technical Report, Census of Population, 2021
Appendix B – The history of sampling in the Canadian census
Sampling was first used in the Canadian census in 1941. A housing schedule was completed for every 10th dwelling. The information from 27 questions on the separate housing schedule was integrated with the data in the personal and household section of the population schedule for the same dwelling. This enabled cross-tabulation of sample and basic characteristics. Also, in the 1941 Census, sampling was used at the processing stage to obtain early estimates of earnings of wage-earners, of the distribution of the population of working age and of the composition of families in Canada. In this case, a sample of every 10th enumeration area across Canada was selected and all population schedules in these areas were processed in advance.
The census of housing was again conducted on a sample basis in 1951. This time, every fifth dwelling (those whose identification numbers ended in a 2 or 7) was selected to complete a housing document containing 24 questions. In the 1961 Census, persons aged 15 years and older in a 20% sample of private households were required to complete a population sample questionnaire containing questions on internal migration, fertility and income. Sampling was not used in the smaller censuses of 1956 and 1966.
The 1971 Census saw several major innovations in the method of census-taking. The primary change was from the traditional canvasser method of enumeration to the use of self-enumeration for the majority of the population. This change was prompted by the results of several studies in Canada and elsewhere (Fellegi 1964; Hansen et al. 1959), which indicated that the effect of the enumerator was a major contribution to the variance of census figures in a canvasser census. Consequently, the use of self‑enumeration was expected to reduce the variance of census figures by reducing the effect of the enumerator and by giving the respondent more time and privacy in which to answer the census questions—factors that might be expected to yield more accurate responses.
The second aspect of the 1971 Census that differentiated it from any earlier census was its content. The number of topics covered and the number of questions asked were greater than in any previous census. Considerations of cost, respondent burden and timeliness versus the level of data quality to be expected using self-enumeration and sampling led to a decision to collect all but certain basic characteristics on a one-third sample basis in the 1971 Census. In all but the more remote areas of Canada, every third private household received the “long questionnaire,” which contained all the census questions. The remaining private households received the “short questionnaire” containing only the basic questions covering name, relationship to head of household, sex, date of birth, marital status, mother tongue, type of dwelling, tenure, number of rooms, water supply, toilet facilities and certain census coverage items. All households in pre-identified remote enumeration areas and all collective dwellings received the long questionnaire. A more detailed description of the consideration of the use of sampling in the 1971 Census is given in Sampling in the Census (Dominion Bureau of Statistics 1968).
The 1976 Census had considerably less content than the 1971 Census. Furthermore, the 1976 questionnaire did not include the questions that cause the most difficulty in collection (e.g., income) or that are costly to code (e.g., occupation, industry and place of work). Therefore, the benefits of sampling in terms of cost savings and reduced respondent burden were less clear than for the 1971 Census. Nevertheless, after estimating the potential cost savings to be expected with various sampling fractions and considering the public relations issues related to a reversion to 100% enumeration after a successful application of sampling in 1971, Statistics Canada decided to use the same sampling procedure in 1976 as in 1971.
Most of the methodology used in the 1971 and 1976 censuses was kept for the 1981 Census, except that the sampling rate was reduced from every third occupied private household to every fifth. Studies done at the time showed that the resulting reduction in data quality (measured in terms of variance) would be tolerable and would not be significant enough to offset the benefits of reduced cost and respondent burden and improved timeliness (Royce 1983). The one-in-five sampling rate was maintained for every census from 1981 to 2006.
In 2011, information previously collected by the mandatory long-form census questionnaire was collected on a voluntary basis, via the National Household Survey (NHS). With this change, every household was required to answer the 10 questions that were contained in the 2011 Census questionnaire, while 30% of households were selected to respond to the NHS. As well, NHS non‑responding households were subsampled for follow-up at a rate of one in three. The increased sampling fraction to 30% was implemented in anticipation of a lower response rate to the NHS. For the 2016 Census, the government reinstated the census long-form questionnaire as mandatory, replacing the NHS. The sampling fraction was changed in 2016 to one in four, compared with one in five for the previous census long-form questionnaire in 2006, to mitigate the risk of the response rate not recovering to its previously high levels.
In 2021, the sample design and sampling fraction of one in four remained nearly identical to those of the 2016 Census. Only a small operational improvement to support a random start to the systematic sampling procedure was made. This was done in an effort to ensure the sample of the two cycles were statistically independent, since dependent samples would have made historical comparisons more difficult to make.
- Date modified: