Aboriginal Peoples Survey (APS)

Detailed information for 1991

Status:

Active

Frequency:

One Time

Record number:

3250

The purpose of the Aboriginal Peoples Survey is to provide data on the social and economic conditions of Aboriginal people in Canada. More specifically, its purpose is to identify the needs of Aboriginal people and focus on issues such as health, language, employment, income, schooling, housing, and mobility.

Data release - June 29, 1993

Description

This is a post-censal survey that collects data on Aboriginal Peoples who identify with their Aboriginal origins or are Registered Indians.

The purpose of the Aboriginal Peoples Survey is to provide data on the social and economic conditions of Aboriginal people in Canada. More specifically, its purpose is to identify the needs of Aboriginal people and focus on issues such as health, language, employment, income, schooling, housing, and mobility.

Subjects

  • Aboriginal peoples
  • Population characteristics

Data sources and methodology

Sampling

This is a sample survey with a cross-sectional design.

The sampling frame used for the off-reserve component of the APS follows a two-stage stratified design. The primary sampling unit (PSU) is made up of enumeration areas consolidated on the basis of specific criteria. The secondary unit is the respondent.

Each domain for which estimates are sought is divided into enumeration areas (EA). Each EA is then divided into two parts: a "single" EA, which includes the single ethnic origin Aboriginal individuals in the EA, and a "multiple" EA, which includes the multiple ethnic origin Aboriginal individuals in the EA.

The sampling frame used for the on-reserve component is random and systematic in design. A list of individuals was drawn for each reserve targeted by the survey to produce a systematic sample.

Data sources

Data collection for this reference period: 1991-06-01 to 1992-01-31

Responding to this survey is voluntary.

Data are collected directly from survey respondents.

Data for the survey were collected through personal interviews with selected persons. For adults, interviews were conducted for the most part with the designated respondent. About 17% of adult interviews were conducted through another household member if the selected adult was absent during the data collection period. For the most part, information on children under 15 years of age was obtained from a parent or guardian; however, some children 12 years of age or older could respond for themselves.

The response rate for Domain 1 was 79%, and 78% for Domain 2. The refusal rate for the APS was low; however contact was not made with 15% of each selected sample.

Error detection

All records in the APS database were then put through a series of checks to verify the validity and consistency of the answers obtained. Missing, incomplete or inconsistent data were considered to be "unknown" or in some cases, the missing data were imputed from other information in the questionnaire.

Data were captured at regional offices of Statistics Canada and then transmitted to the central office of Statistics Canada for processing, and the questionnaires were sent to Ottawa. The names and addresses of individual respondents were not entered into the APS database.

Estimation

In a survey such as the APS, each respondent in the sample is representative of a sub-group of the population studied. Each record in the database is therefore given an initial weight corresponding to the number of persons it represents. The initial weight is adjusted to compensate for non-responses and for discrepancies between the characteristics of the sample and those of the target population. The results of the survey are then multiplied by the numeric weight to obtain an estimate for what the response would be of an entire population. The results of the APS were weighted to be representative of the census Aboriginal population, excluding those reserves and communities that were incompletely enumerated.

All statistics obtained from the Aboriginal Peoples Survey (APS) database are estimates based on a probability survey carried out with a sample of Canada's Aboriginal population. Consequently, these statistics can be subject to two different types of errors: sampling errors and non-sampling errors.
Sampling errors can be defined as errors caused by the analysis of a sample rather than a census of the entire population carried out under the same conditions. In the context of a survey such as the APS, we can estimate this error using the survey data. The measure of error used is called the coefficient of variation (CV) of the estimate, which is the standard error of the estimate divided by the estimate itself.
For the purpose of the survey, when the coefficient of variation (CV) of an estimate is higher than 33.3%, this is considered too unreliable to be published and the symbol " -- " is printed in the corresponding cell of the data table. This symbol is also used to indicate that publication of the data in that cell would violate confidentiality rules. When the CV of the estimate is 16.6% to 33.3%, the corresponding estimate is accompanied by the symbol " * " in the table. These estimates should be used with caution to support a conclusion. All estimates where the CV is lower than 16.6% can be used unconditionally.

All other types of errors (i.e. coverage, response, processing and non-processing errors) are categorized as non-sampling errors. It is generally difficult to identify and evaluate the effect of some of these errors.
Coverage errors occur when there are differences between the target population and the population sampled. In the case of the APS, integration of the survey with the structure of the 1991 Census greatly reduced this type of error. However, because the APS sample is selected from those who participated in the census, the APS information is unavailable for those communities that were incompletely enumerated in the census. According to the 1991 Census data, 78 Indian reserves and settlements were incompletely enumerated in the census, and 181 Indian reserves and settlements were incompletely enumerated in the APS even though they had participated in the census. Other persons not covered by the APS include Aboriginal people living in collective dwellings, institutions, military camps, and overseas.

A response error occurs when the respondent misunderstands a question and the interviewer records an incorrect answer. Several procedures were taken to minimize this type of error, including interviewer training, respondent follow-up, and extensive response edits.
Processing errors may occur at various stages including coding, data capture, and imputation. Quality control measures are applied to every stage of the data processing in order to minimize this type of error.
Any statistical survey is affected by a certain percentage of non-response among the selected sample. Non-response is said to be total when, for any reason, a selected person could not be interviewed for the survey. Non-response is partial when only part of the questionnaire is completed. Non-response errors depend on the type and degree of differences that may exist between the characteristics of the sample of respondents.

Quality evaluation

Coverage errors occur when there are differences between the target population and the population sampled. In the case of the APS, integration of the survey with the structure of the 1991 Census greatly reduced this type of error. However, because the APS sample is selected from those who participated in the census, the APS information is unavailable for those Communities that were incompletely enumerated in the census. According to the 1991 Census data, 78 Indian reserves and settlements were incompletely enumerated in the census, and 181 Indian reserves and settlements were incompletely enumerated in the APS even though they had participated in the census. Other persons not covered by the APS include Aboriginal people living in collective dwellings, institutions, military camps, and overseas.

Non-response is said to be a total when, for any reason, a selected person could not be interviewed for the survey. Non-response is partial when only part of the questionnaire is completed. Non-response errors depend on the type and degree and differences that may exist between the characteristics of the sample of respondents and those of the sample of non-respondents. In principle, the greater the differences, the greater are the effects on the accuracy of the estimates.

The rates of response of the APS (79% for Domain 1 and 78% for Domain 2) were acceptable. Although these rates may vary significantly from one settlement or reserve to another, various measures were taken to lessen the bias caused by the number of non-responses. For example, in Domain 1, a correction for non-response was made by community, age group, and sex. In the data tables, non-responses to specific questions are identified under the heading "unspecified".

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects that could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.

Documentation

Report a problem on this page

Is something not working? Is there information outdated? Can't find what you're looking for?

Please contact us and let us know how we can help you.

Privacy notice

Date modified: