The historical comparability of Aboriginal data has been affected by legislative changes such as the amendment to the Indian Act of Canada known as Bill C-31, in 1985, and changes in the social environment resulting from legal milestones such as the Powley case1 and their impact on reporting changes related to Aboriginal self-identification. These reporting changes refer to people changing, from one census to the next, the reporting of their Aboriginal affiliations from non-Aboriginal to Aboriginal affiliations as well as from Aboriginal to non-Aboriginal affiliations. (Guimond, 2003; Siggner and Costa, 2005 ethnic mobility).2,3
As well, over time there have also been some changes in the questions asked, their wording, format and examples used in the census. However, three of four Aboriginal questions in the census have remained unchanged in the past three censuses (1996, 2001 and 2006) with the exception of examples used in some questions, and the changes in the ancestry question.
The following are the principle factors that can affect the comparability of Aboriginal census data over time:
Some Indian reserves and settlements did not participate in the census as enumeration was not permitted or it was interrupted before completion. In 2006, there were 22 incompletely enumerated reserves or settlements, down from 30 in 2001 and 77 in 1996.
In 1996, an estimated 43,566 persons were missed as a result of incompletely enumerated reserves and settlements. This decreased to 34,500 in 2001, and increased to an estimated 40,115 in 2006.
Data showing changes between censuses can be adjusted to include only reserves enumerated in the censuses being compared. This approach was followed in the dissemination report on the Aboriginal population from the 2006 Census Aboriginal Peoples in Canada in 2006: Inuit, Métis and First Nations, 2006 Census.
For the lists of incompletely enumerated reserves, please refer to the 1996 list, the 2001 list and to the 2006 list. Note that each of these links provides a table listing the incompletely enumerated reserves for that particular census and also provides the population counts for previous censuses, when available.
Undercoverage occurs when persons or dwellings are missed during enumeration. Persons can be missed when dwellings are missed or when there is a misunderstanding about whom to include on the questionnaire.
In 2006, it is estimated that the net undercoverage of the population living in participating Indian reserves and settlements is 40,623 persons or 10.6%, compared to 2.7% of net undercoverage for the total Canadian population in the 2006 Census.
The undercoverage of the off reserve Aboriginal population for 2006 is not available.
Presently, there are four questions used in the census to identify Aboriginal peoples: ethnic origin/ancestry; Aboriginal identity; registered or treaty Indian; and member of an Indian band or First Nation. With the exception of the ethnic origin/ancestry question, the other questions have remained unchanged in the 1996, 2001 and 2006 censuses but there were minor changes to examples used in the questions. In terms of question wording or format, the Aboriginal data from the 1996, 2001 and 2006 censuses are comparable as no changes have been made to these questions. There are other factors which can affect comparability of the data such as changes in the social environment and these are discussed below. As well, a brief overview of the question changes prior to 1996 is included to provide a larger context to understanding the data over time.
An ethnic origin question has been asked in the census since 1871. However, the question has undergone several changes over the years. The areas that underwent changes involved the criteria used to determine ethnic origin or ancestry, the terminology used to designate Aboriginal peoples, as well as changes to the question itself, and to the way the data have been collected. For more details see: How Statistics Canada Identifies Aboriginal Peoples.
The concept of ethnicity is fluid and is probably the most complex concept measured in the census. Respondents' understanding or views about ethnicity (ancestry) and awareness of their family background affect the reporting of ethnicity (ancestry) from one census to another. Increasing intermarriage among various groups has led to an increase in the reporting of multiple origins (ancestries), which has added to the complexity of the ethnic (ancestry) data. Furthermore, reporting patterns are affected by changes in the format and wording of the question, as well as changes in the examples provided on the question.
Thus the historical analysis of ethnic origin data is affected by these factors, as well as by changes in the social environment at the time of the census.
An Aboriginal identity question was first included in the census long form questionnaire in 1996 4. This same question was also asked in the 2001 and 2006 censuses.
Note that in the 1996, 2001 and 2006 censuses, the total Aboriginal identity counts were derived using three questions (questions 18, 20 and 21), that is by including those persons who reported that they identify with at least one Aboriginal group, i.e., North American Indian, Métis or Inuit and/or those who reported being a Registered Indian or Treaty Indian as defined by the Indian Act of Canada, and/or those persons who were members of an Indian Band or First Nation.
In terms of question wording or format, the Aboriginal identity data from the 1996, 2001 and 2006 censuses are comparable as no changes have been made to this question.
However, caution should be exercised in analyzing trends in Aboriginal identity based on data from previous censuses. Although the data have not been affected by changes in the question, the growth in the census counts of the Aboriginal identity population has been affected by both demographic factors (birth, deaths and migration) and non-demographic factors such as reporting changes in those identifying as Aboriginal and by the relative number of incompletely enumerated Indian reserves and settlements.
The information on Indian legal status was collected in the census for the first time in 1981 as a sub-question of the ethnic origin question. It has been collected as a separate question since 1991. The decision to separate the concept of ancestry from that of Indian status was due to recognition that the latter is one's legal status in relation to the Indian Act, rather than one's ancestral origins.
Furthermore, the wording of the 1996 Census question on Registered or Treaty Indian differed slightly from the one used in previous years. Prior to 1996, the term 'Treaty Indian' did not appear in the question. It was added in 1996 at the request of individuals from the Western provinces, where the term 'Treaty Indian' is more widely used than the term 'Registered Indian.' It was felt that this change would make the question more understandable to those Registered or Treaty Indians living in the west.
In terms of question wording or format, the Registered or Treaty Indian data from the 1996, 2001 and 2006 censuses are comparable.
The Registered or Treaty Indian counts from the census are different from those obtained from the Indian Register which is maintained by Indian and Northern Affairs Canada, the details of these differences are covered in section 6.2 Comparability with Indian Register.
This question asks respondents if they are members of an Indian band or First Nation and if yes, to indicate the name of the Indian band or First Nation.
Information on membership of an Indian band was first collected in 1991 as a sub-element of the Registered Indian question. In the first part of the question, respondents were asked about Indian status, while the second part of the question dealt with Indian band membership.
In 1996, a separate question was developed to collect data on Indian band/First Nation membership, many Indian bands having elected to call themselves a First Nation and having changed their band name to reflect this.
In addition, with the 1985 amendment to the Indian Actof Canada (Bill C-31), many Indian bands exercised the right to establish their own membership code, whereby it was not always necessary for a band member to be a Registered Indian.
The historical comparability of Indian band or First Nation membership data has thus been affected by these factors, and the data should be used with caution. Data on Indian band membership are also available from the Indian Register and there are differences in the Indian band population counts from the census and the Indian Register, due to the incomplete enumeration of some Indian reserves as well as methodological, conceptual and collection differences between the census and the Indian Register.
The Census of Population from Statistics Canada and the Indian Register maintained by Indian and Northern Affairs Canada are the two principle sources of demographic data on the registered (or status) Indian population. The registered Indian population is a component of the Aboriginal population (derived variable). Registered Indian refers to those persons who reported they were registered under the Indian Act of Canada, irrespective of their Aboriginal affiliation.
The census is a self-administered survey that is taken every 5 years. It is a snapshot of the population at a moment in time--most recently on May 16, 2006. The Indian Register is a continuous statutory administrative file based on the registration of individuals who meet specific criteria as defined by the Indian Act.
The different purposes, as well as methodological and conceptual differences between the Indian Register and the census often result in differences in counts for the registered Indian population from each of the data sources.
The recorded count of registered Indians from the census, on May 16, 2006, is 623,780 compared to 755,940 from the Indian Register, on May 31, 2006.
The census count is affected by:
The census does not collect data on registered Indians living in institutions (for example, hospitals, senior citizens' homes, jails, shelters, etc.); as well as registered Indians living outside of the country on census day.
The Indian Register count is affected by:
In the Indian Register there is a reporting lag between the occurrence of an event and it being reported and recorded in the Indian Register.
Adjustments to both the Census and the Indian Register will result in a much smaller difference in counts between the two sources.