History of the Census of Canada

Canada's early censuses

The first census in Canada was initiated by Intendant Jean Talon in 1666. The census counted the colony's 3,215 inhabitants and recorded their age, sex, marital status and occupation. In light of the need for information to help plan and develop the Colony of New France, Talon did much of the data collection personally, visiting settlers throughout the colony.

The years leading up to 1871 (Canada's first census) saw a long string of 98 colonial and regional censuses. During this period, notably in enumerations between 1710 and 1760, there was a particular need to collect information on housing stock and armaments such as muskets and swords owned by households. Previous censuses had been more concerned with raising taxes or armies and assessing resources.

Later, as other needs arose, questions were added on livestock, crops, buildings, churches, grist mills and firearms. In 1765 for example, priorities included assessing the balance between Catholics and Protestants, and recording the number of Acadians, settlers, Aboriginals and Blacks. As a result of this census, new variables that included questions on race, religion and ethnic origin were introduced.

The first census after Confederation

The first census of Canada was taken in 1871. According to The Census Act of May 12, 1870, census-taking was to take place no later than May 1st. Under Section 8 of the Constitution Act of 1867 (formerly the British North America Act), a census was to be taken in 1871 and every tenth year thereafter. This first census of the dominion following confederation in 1867 counted the population of the four original provinces of Nova Scotia, New Brunswick, Quebec and Ontario. Its main goal was to determine appropriate representation by population in the new parliament. Since 1871, decennial census data have provided the cornerstone for representative government.

In 1871 the questionnaire covered a variety of subjects, and asked 211 questions on area, land holdings, vital statistics, religion, education, administration, the military, justice, agriculture, commerce, industry and finance. Information was collected in tabular form on population, houses and other buildings, lands, industries and institutions. The population questions included the age, sex, religion, education, race and occupation of each person. Not every household answered all 211 questions.

A new tradition

The 1871 Census began collecting : information on the ancestral origins of all Canadians, including Aboriginal persons, was recorded. Following data collection, between 35 and 50 clerks, unaided by machines of any kind, compiled results that were published in five bilingual volumes in 1873.

It is interesting to look at the use of terms in the census at this time. For example, the only options for “marital status” were married, widowed or other. Today there are five categories: legally married, separated but still legally married, divorced, widowed, or never legally married. Similarly, from 1871 to 1911, the census asked questions on “infirmities”. Respondents were asked to indicate whether members of their household were “blind, deaf, or simple-minded.” These questions were dropped from the 1921 and subsequent censuses.

Beginning in 1906, the prairie provinces of Manitoba, Alberta and Saskatchewan began to take a separate census of population and agriculture every five years to monitor the growth of the west. Since 1956, the Census of Agriculture and the Census of Population have been taken together every five years across the entire country.

Canada's growing population and census

Two major changes were made to the census in 1881. All census takers were required to take an oath of secrecy, a pledge still required today. The census was extended to include British Columbia, Manitoba and Prince Edward Island. Also, a housing question was revised to include “wigwams and tents” in the census of the Northwest Territories.

In 1891, the population was prepared for the census enumerator's visit through announcements in newspapers and from pulpits.

The 1901 Census was truly a huge undertaking in that its content grew from nine questionnaires and 216 questions in 1891 to 11 questionnaires and 561 questions—though again, not all households answered all questions. The additional questions on religion, birthplace, citizenship and period of immigration supplemented the existing question on ethnic origin. By this time, the population of Canada was 5,371,051, and Montréal was the most populous city, with 267,730 residents.

The 1911 Census had 13 questionnaires with 522 questions. Some language questions were added, but results were never released. The census showed that the general house-to-house census enumeration should no longer include the detailed fishery questionnaire with questions such as the quantity, kind, and value of catch; and the number and type of boats, gear and equipment. Instead, a special form was used in specified fishing areas of Canada.

Creation of the Bureau of Statistics

In 1912, responsibility for the census shifted from the Ministry of Agriculture to the Ministry of Trade and Commerce. Six years later, the Dominion Bureau of Statistics was created.

With the introduction of household and business surveys, the number of census questionnaires decreased since there were now other ways to obtain information on Canada’s population. The 1921 Census had only five questionnaires; however, the number of questions had increased to 565—up from 522 in 1911. The five questionnaires were population; agriculture; animals, animal products and fruits not on farms; manufacturing and trading establishments; and a supplemental questionnaire for persons who were blind and deaf. The population questionnaire contained only 35 questions, with those on “insanity” and fertility having been dropped. A new question recorded the birthplaces of the father and mother of each individual.

By 1921, field work, which consisted primarily of canvassers completing forms, included the training of commissioners and enumerators and the checking and follow-up of census questionnaires for discrepancies or omissions. A special staff, varying from 120 to 350 at peak times, compiled the census results using mechanical tabulation methods.

Compilation and tabulation for the 1931 Census were still carried out with mechanical equipment. However, a new sorter-tabulator developed by Fernand Bélisle, an employee of the Dominion Bureau of Statistics, made production 50 times faster than was previously possible. The device allowed a whole data card to be read at once rather than one column at a time.

The changing census

In the 1931 Census, questions were asked to gauge the extent and severity of unemployment and to analyze its causes. The 1941 Census, developed during the Depression and conducted during the Second World War, evaluated data for “social areas” in Winnipeg and Vancouver. This was the first census that linked the urban poor with a view to developing urban planning. It also paved the way for fiscal arrangements between the federal and provincial governments in the form of equalization payments and transfers for health care and post-secondary education.

The 1941 Census marked a number of innovations. The “description of home” questions were supplemented with a questionnaire of housing which sought extensive information on dwellings, including the type, facilities, tenure, rent, value, mortgages, etc. The questionnaire was addressed to every tenth household and was the first use of sampling by the census. This method of gathering detailed data from a sample of households rather than all households proved to be effective, yielding high-quality data while reducing costs and response burden. In 1951, the sample was expanded to one household in five in order to obtain greater geographic detail. Since then, this ratio has been used in all decennial censuses except the Census of 1971, when a one in three sample was used.

Another major innovation was the inclusion of questions on fertility, a topic considered too sensitive for the 1921 and 1931 Censuses. Fertility was correlated to earnings, schooling and other significant characteristics of the family.

The postwar years

The 1951 Census, held two years after Newfoundland became part of Canada, marked Canada's first census as a nation of 10 provinces and two territories. The content of the population and housing questionnaire covered name, sex, age, marital status, relationship to “head” of household, and the structural type and tenure of dwelling. This census also provided information for small areas such as counties, municipalities, cities, towns, etc. These data proved beneficial to users for a number of purposes including determining grants and subsidy allowances, studying social and economic problems, and conducting market research.

For the 1951 Census, enumerator “mark-sense” technology used for the population and housing questionnaires greatly reduced processing time and costs. This method recorded respondent information by means of special markings in designated positions on the enumeration document. Punch cards could then be generated by a document punch without the intermediate step of manual punching. Subsequent processing of the cards was carried out on a high-speed tabulator.

In 1956, the first nationwide quinquennial census was conducted and enjoyed the distinction of being the first to use television in its publicity program. Rapid growth in population and agriculture indicated the need for benchmarks at five-year intervals in order to provide a more accurate basis for annual estimates. The major challenge of a complete Census of Canada was to develop a simplified questionnaire restricted to the essentials so as not to exceed the allocated budget.

Enumeration was carried out by the Bureau's eight regional offices. The enumeration process included a postal check of 60% of Canadian households. Letter carriers in 170 urban areas checked enumerators' address listings against their own records. A subsequent follow-up of missed addresses added over 40,000 persons to the total population count.

The only new question in the 1961 Census concerned the level of education of household members. A 20% sample of households was used to collect additional information on internal migration, fertility, and income characteristics; the same sample was used for the housing census. The 1961 Census planning committee decided that computer facilities would be used for processing respondent information collected in the field on “mark-sense” documents and then transferred by an optical document reader to magnetic tape for later computer processing.

The 1966 Census was the second mid-decade census of the entire country. The content of the population and housing questionnaire was similar to that of 1956. The agricultural questionnaire contained 138 questions, compared with 251 in 1961.

The contemporary census

By 1971, the Dominion Bureau of Statistics had become Statistics Canada. This year also marked the 100th anniversary of the first census of Canada. Under the new Statistics Act, it became a statutory requirement to hold censuses of population and agriculture every five years.

The 1971 Census will be remembered for the introduction of more methodological innovations than any of its predecessors. Self-enumeration, whereby respondents complete their own questionnaire, was introduced, representing a major change in collection method. Every census up to and including the 1966 Census had been conducted by interview.

Another important innovation was the reverse record check which sought to determine under-coverage using a sample of the census.

The content of the questionnaire was greatly enlarged. New questions on vocational education, language spoken at home, full-time and part-time employment, place of work, and farm income were added; this census was the last to ask questions on military service. In the agricultural census, environmental concerns gave rise to new questions on topics such as the use of pesticides and fertilizers.

Two questionnaires were used in 1971. The “short” form distributed to two-thirds of Canadian households covered the basic population questions and nine housing questions. The “long” form, distributed to the remaining third, contained the same questions as the short form with the addition of 20 housing questions and 30 socio-economic population questions. The Census of Agriculture questionnaire contained 199 questions, down from 251 in 1961.

The larger population in 1971 and the increased complexity of the population and housing census necessitated an increase in the number of census enumerators—from 30,000 in 1961 to 41,000 in 1971. A “drop-off, mail-back” questionnaire was used in large urban areas; in rural areas, questionnaires were both dropped off and picked up by census enumerators. The remainder of the population—those living in northern regions, coastal outports, institutions, etc., were enumerated by traditional face-to-face interviews.

Similar to the 1956 and 1966 Censuses, a mini-census (mid-decade) with fewer questions was conducted again in 1976. The 1986 Census broke the pattern established in 1956 of alternating full and mini-censuses by repeating most of the questions asked in the full census of 1981. Also in 1986, a large post-censal survey on health and activity limitations was held for the first time.

The term “head”, which previously referred to the husband in a husband-wife family or the parent in a lone-parent family, was changed in 1976 for husband-wife families to refer to either the husband or wife. The reference to “head” was dropped altogether in the 1981 Census. In 1991, a question on “common-law” relationships was included on the census questionnaire for the first time. In 1996 more new questions were added including Aboriginal identity, population group, household activities or unpaid work, and mode of transportation to work.

Materials for the Censuses of 1986 and 1991 were available in alternate formats such as braille, audiocassette, and large print. The 1996 Census was translated into 49 non-official languages, 12 of which were Aboriginal languages, and was available in the same alternate formats as well as diskette. Eighty percent of Canadian households completed the short questionnaire with seven questions, while the remaining 20% completed the long questionnaire with 55 questions.

In 2001, two new questions were introduced, one on birthplace of parents, the other on language of work. The definition of common-law couples was changed to include members of the opposite-sex or same-sex living together as a couple, but who are not legally married to each other.

In the 2006 Census, forms were delivered by Canada Post to about 70% of households. The remaining 30% received the form from an interviewer as in previous censuses.

For the first time, the census counted same-sex married couples, reflecting the legalization of same-sex marriages for all of Canada as of July 2005.

A new question seeking permission for Statistics Canada to use data from income tax records to lower respondent burden was added to the long form. As well, questions on education were re-worded to improve response quality, including a new question on location of study. Both the short and long forms contained a new question asking whether the respondent would permit Statistics Canada to make their personal information public in 92 years for historical and genealogical research.

This was the first time that households across the country were offered the option of completing their questionnaire online. In fact, 18.5% of Canadian households used the online option. This new method place d Canada at the forefront of census-taking.

The 2011 Census

In early May, households received either a letter or a questionnaire package. The letter replaced the traditional paper questionnaire to encourage respondents to complete the questionnaire online. The Internet response target was increased from 20% in 2006 to 40% in 2011.

All households were asked to complete ten questions on basic topics such as relationship to Person 1, age, sex, marital status, language and consent to future release of personal information.

The 2011 Census also included a series of mail and telephone reminders to encourage households to complete their questionnaires during May 2011. This was to ensure better quality data and reduce costs associated with follow-up.

For the first time, the questions normally asked on the census long form were asked in the National Household Survey. Questionnaires for this voluntary survey were distributed approximately four weeks after the census.

And so, in 2011 we will have travelled through 345 years of census history in Canada. We have grown from 3,215 inhabitants to a nation of about 34 million. Census-taking in Canada has adapted and changed as we have grown from small groups of settlers to a nation of urban metropolises.