Aboriginal groups and geography

Projections have been prepared for persons who identify with any of the following groups: the North American Indian population, the Métis or the Inuit. Geographical detail for each group is different and depends on the group's size. All three groups were projected separately for each of the ten provinces and three territories. However, the subprovincial and subterritorial level shown for the three groups varies. For the North American Indians, future numbers were calculated for the urban parts of all census metropolitan areas (CMAs), urban areas outside CMAs, rural areas, and reserves. For the Métis, places of residence were grouped into urban parts of CMAs, urban areas outside CMAs, and rural areas, which also comprise reserves. Because of their relatively small size, the Inuit population was projected separately for urban and rural locations only. Figure 1.1 presents the geographical details for all three Aboriginal groups.

Figure 1.1
The Aboriginal groups and geography

Approach used

Projections of the Aboriginal populations in Canada were produced using the traditional regional cohort component approach. According to this model (Figure 1.2 ), the size of the group at time (t+1) is the result of demographic events which happened in this group during the previous year. Those events could be: birth given by a woman; death of a member of the group; change of place of residence involving crossing or not crossing regional boundaries. Also, this model typically considers changes of residence involving country borders - international migration of persons. We assumed here that international migration of persons identifying with the three Aboriginal groups was of negligible importance to the change of their size ¹. In the past, changes in ethnic reporting contributed significantly to growth of the Métis population and the North American Indian population, specifically in areas outside of reserves. Although reporting of self-identity of Aboriginal identity could change over time, this element was not considered in these projections. Such changes in reporting patterns are too volatile to be projected. Therefore, the projections of the Aboriginal populations are based on demographic components of change, for the populations related to fertility, mortality and internal migration during the projection period.

Figure 1.2
Projection model - the cohort component method

Starting point

The project considers the population who, in the 2001 Census, declared that they had an Aboriginal identity and/or were Registered or Treaty Indians ² and/or were members of an Indian Band. There were 952,900 people with at least one Aboriginal identity ³, most of them identifying with only one Aboriginal group (946,200). A small number of people, less than 6,700, also indicated belonging to more than one Aboriginal group. In addition, 23,400 people just reported their status as a Registered or Treaty Indian or a member of an Indian Band. Together, this gave a base population of 976,300 people for our projections.

Before using this population as the base for projections, two modifications to the 2001 Census counts were made. First, it was necessary to assure that those persons who indicated that they belonged to more than one Aboriginal group were considered only once in the model, and that 23,400 people who didn't identify Aboriginal were also included. This modification did not change the overall size of the Census Aboriginal population (976,300), but introduced small changes to the size of the North American Indians as 635,700 (instead of 608,800), the Métis as 295,100 (instead of 292,300) and the Inuit as 45,500 (instead of 45,100).

The second modification of the census data was to correct for situations when the census didn't count all those who should have been counted. Contrary to the first modification, this change increased the size of the total Aboriginal population and that of specific groups. The population living on-reserve was also adjusted to take into account incompletely enumerated reserves. In each census, on some Indian reserves and settlements, enumeration is not permitted or is interrupted before it is completed. Estimates of the number of people not counted were made and added to the census number of persons living on reserves. In 2001, there were 30 such reserves, with an estimated total population of 31,000 ⁴. Also, during each census enumeration, there are people who cannot be counted for various reasons (their dwelling was missed, they were not included on the census questionnaire, etc.) or who are counted more than once. Adjustment for these situations was made for the Aboriginal populations living on enumerated reserves and outside the reserves ⁵.

Finally, after all modifications, the starting point for the published projections was 1,066,500 Aboriginal people in 2001, an increase of over 9% (or 90,200) from the census number. The size of the North American Indian group was modified the most, from 635,700 to 713,100, a difference of 77,400 or 12.2%. The largest contribution to the increase of this group came from the adjustment of the on-reserve population for net undercount. For the Métis, the impact of adjustments was smaller: their number increased from 295,100 to 305,800, an increase of 10,700 or 3.6%. And, the Inuit population showed a modest increase of 2,100 (or 4.6%), from 45,500 to 47,600. In the remainder of this section let us now discuss the data sources and methods used to derive information on fertility, mortality and migration that were used in the projection model.

Estimating births

Fertility has a direct impact not only on the size of the projected population but also on its composition. The number of births during the projected period depends on several factors, such as the number of women who could be mothers, their age, their membership in a particular Aboriginal group and their place of residence. The most comprehensive source on births is birth registration data compiled by the Health Statistics Division of Statistics Canada. Unfortunately, this dataset does not provide information on the fertility of populations defined by their Aboriginal identity. In fact, direct data on the fertility of the Aboriginal peoples are sparse. Indian and Northern Affairs Canada (INAC) compiles data on the First Nations (Registered Indian) population through the Indian Register, but there is no administrative source specific to the Non-Status Indian, Métis or Inuit populations.

As a result, for this projection model, estimates of the total fertility rate had to be developed for the North American Indians, the Métis and the Inuit. To maintain internal consistency and to avoid the situation where data for each Aboriginal group would be taken from a different source, birth data from the Indian Register were not used. Instead, we used the 2001 Census data and an indirect technique, known as the "own-children method", to derive the fertility level of the three Aboriginal groups, for regions and areas of their residence ⁶. Also, we calculated the mean and modal ages of fertility required by the parametric module of fertility in our projection model.

For the North American Indian and the Métis populations, the provinces and territories were grouped into two regions: a high-fertility region and a low-fertility region (Figure 1.3 ). This approach was used to avoid calculation of fertility rates based on small numbers, as some Aboriginal groups in particular provinces and territories were represented by relatively small populations. The grouping was based on the level of fertility for the total Aboriginal population in Canada estimated for two periods: 1991 to 1996 and 1996 to 2001. During both periods, total fertility rates of the Aboriginal peoples in Manitoba, Saskatchewan, the Yukon Territory, the Northwest Territories and Nunavut were higher than the national average; thus, these provinces were combined as a high-fertility area. Conversely, the low-fertility area consisted of Newfoundland and Labrador, Prince Edward Island, Nova Scotia, New Brunswick, Quebec, Ontario, Alberta and British Columbia.

Figure 1.3
Regions of fertility

For the Inuit, fertility levels were estimated separately for two provinces - Newfoundland and Labrador, and Quebec - and for one territory - Nunavut. The rest of the regions had the same level of fertility.

Estimating deaths

The studies conducted so far demonstrated that regardless of gains in the past, the Aboriginal mortality remains higher than mortality of the total Canadian population. There were also differences in mortality within the Aboriginal population - among the three groups and with respect to place of residence. Projecting the three Aboriginal groups required the number of deaths for each cohort born before 2001 and those who would be born during the projection period. In our model, the baseline mortality of 2001 was estimated through life expectancy at birth for the North American Indians, the Métis and the Inuit. Values of life expectancy were used to provide probabilities of surviving one year for women and men at each single year of age for each geographical area and place of residence (on a reserve, in a CMA, in an urban non-CMA area, and in a rural area). The challenge lay in the fact that the mortality of the Aboriginal population was largely unknown due to the lack of data covering all groups.

The INAC Register is one of the few sources of direct data on mortality for the Aboriginal peoples; however, it provides information on deaths of members of the First Nations only. It was used to estimate and project the mortality of peoples registered under the Indian Act. As in the case of births, the statistics on deaths compiled for the Statistics Canada Mortality Database did not allow for the separation of the Aboriginal and non-Aboriginal peoples. In the absence of information, other than for Registered Indians, in our projections we used a proxy to help estimate life expectancy for all North American Indians, the Métis and the Inuit. We combined data from several sources: life tables for Registered Indians estimated for 2000 and projected for 2001 to 2021 by Statistics Canada Demography Division; life tables for Nunavut and the total Canadian population estimated by Statistics Canada Health Statistics Division for 2001 and projected by Demography Division into 2026; and the 1996 mortality estimates for Canada's urban areas by neighbourhood income prepared by the Health Analysis and Measurement Group of Statistics Canada.

The relationship between socio-economic conditions and health, and, hence, mortality, is widely recognized. We exploit this finding by proposing that the mortality of Non-Status Indians is inferred from that of Registered Indians and mortality of those Canadians who live in neighbourhoods with the lowest income ⁷. The overall mortality for the North American Indians in 2001 would be a combination of the mortality of Registered Indians and Non-Status Indians. Métis mortality is assumed to be similar to that derived for the Non-Status Indians. And finally, Inuit life expectancy is estimated on the basis of mortality registered through vital statistics collected from Nunavut and results of a study done by l'Institut de la statistique du Québec ⁸.

Estimating migrants

The third component of growth considered in our projection model was internal migration. Internal migration has very little effect on the size of the projected population at the Canada level. The national effect is related to possible changes in future behavior of persons as they migrate among geographical areas with different levels of fertility and mortality. This component, however, plays a very important role when projecting the geographical concentration of the Aboriginal groups. In the model, we made a distinction between changes of residence involving a provincial or territorial boundary change and a change within the same geographical region that resulted in a change in the place of residence, such as on-reserve, CMA, urban outside CMA, and rural areas. The former are known as interregional migration and the latter as intraregional migration.

A number of studies demonstrated that, contrary to the total Canadian population, interregional migration was a minor component of regional growth among the Aboriginal populations and continued to represent a fairly small factor in changes to the geographical distribution of the Aboriginal population of Canada. In the long term, however, such a migration pattern could contribute to shifts in the distribution of this population. Intraregional migration had a relatively larger effect on the concentration of the Aboriginal populations. This effect was not only significant but had been evolving as illustrated by the analysis of past migration patterns and levels of the three Aboriginal groups based on census data.

Projection of the number of migrants in our model used census data and took into account the differences in propensity to migrate specific to the Aboriginal group. Within the Aboriginal group, estimates of migrants were prepared by region (provinces and territories) and place of residence. We calculated the number of those who left a particular area (out-migrants) and the number of persons who arrived into this area from any other location (in-migrants). Then, regardless of residence, migrants were assigned the general demographic profile of specific type of migrants. We had four different profiles typical for those leaving reserves, migrating from CMAs, moving out of urban non-CMA areas or leaving their rural abodes. For in-migrants, we also had four age-sex specific profiles.