Aboriginal Peoples Survey (APS)

Detailed information for 2017

Status:

Active

Frequency:

Every 5 years

Record number:

3250

The purpose of the Aboriginal Peoples Survey (APS) is to provide data on the social and economic conditions of First Nations people living off reserve, Métis and Inuit, aged 15 and over, in Canada.

Data release - November 26, 2018

Description

The 2017 APS is a national survey of First Nations people living off reserve, Métis and Inuit aged 15 years and over. The 2017 APS represents the fifth cycle of the survey and focuses on transferable skills, practical training, use of information technology, Aboriginal language attainment, and participation in the Canadian economy.

The 2017 APS collected unique and detailed data on employment, education, and health which are not available from any other source. For example, although the 2016 Census of Population collected data on certain aspects of labour market participation, the 2017 APS addresses additional topics such as job satisfaction, multiple employment, past job attachment, and willingness to move to improve career opportunities.

The APS provides key statistics to inform policy and programming activities aimed at improving the well-being of Aboriginal Peoples. It is a valuable source of information for a variety of stakeholders, including Aboriginal organizations, communities, service providers, researchers, governments, and the general public.

The survey is carried out by Statistics Canada with funding provided by three federal departments: Crown-Indigenous Relations and Northern Affairs Canada & Indigenous Services Canada (formerly Indigenous and Northern Affairs Canada), and Employment and Social Development Canada.

Subjects

  • Aboriginal peoples
  • Aboriginal society and community
  • Education, literacy and skills
  • Health and well-being
  • Households, housing and environment
  • Languages and cultures
  • Population characteristics
  • Work, income and spending

Data sources and methodology

Target population

The target population of the 2017 APS is composed of the Aboriginal identity population of Canada, 15 years of age or older as of January 15, 2017, living in private dwellings excluding people living on Indian reserves and settlements and in certain First Nations communities in Yukon and the Northwest Territories (NWT). The concept of "Aboriginal identity" refers to those persons who reported identifying with at least one Aboriginal group, namely, First Nations (North American Indian), Métis or Inuit; those who reported being a Status Indian (Registered Indian or Treaty Indian, as defined by the Indian Act of Canada); or those who reported being a member of a First Nation or Indian band.

The APS selected its sample from reported answers to the 2016 Census of Population long-form questionnaire. More precisely, the APS sample was selected from individuals who answered "Yes" to one of the three census questions defining the identity population (questions 18, 20 and 21) or those who reported Aboriginal ancestry to question 17. Although not part of the 2017 APS target population, some individuals with Aboriginal ancestry who did not report Aboriginal identity were still sampled, since past survey experience indicates that nearly one-third of these individuals will report an Aboriginal identity on the APS. Therefore, unlike the target population, the sampled population (or survey population) was composed of both the identity population and the Aboriginal ancestry-only population.

Instrument design

Although the 2017 APS was designed to be thematic, it is based on previous cycles of the APS which were developed in collaboration with the national Aboriginal organizations. Following the release of data from the 2012 APS, a content review was conducted to ensure the future relevance of existing APS questions to key stakeholders and to identify any potential data gaps. The review brought together expertise from a diverse group of researchers and subject matter experts, both from within and outside of Statistics Canada. New survey questions were developed and added to the 2017 APS questionnaire in order to place greater emphasis on the themes of economic participation and education.

As was done in 2012, the questions in the 2017 APS were designed for use in a Computer Assisted Interviewing (CAI) environment which incorporates many features that serve to maximize the efficiency and quality of data collection. CAI allows for more complex questionnaire flows as well as on-line edits which identify any logical inconsistencies so that interviewers can correct these with the assistance of respondents at the time of the interview. Two computer assisted interview questionnaires were developed for this survey: a Computer Assisted Telephone Interview (CATI) and a Computer Assisted Personal Interview (CAPI).

Qualitative testing of the survey questionnaire was carried out by Statistics Canada's Questionnaire Design Resource Centre (QDRC) with the help of First Nations people, Métis, and Inuit across Canada. Adjustments were made to question wording and flows based on those results. Question wording adheres as closely as possibly to questions established by the Harmonized Content Committee at Statistics Canada. This increases opportunities to compare responses between Statistics Canada surveys.

Sampling

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

Survey frame:
The APS sample was selected from the 2016 Census of Population long-form respondents who reported an Aboriginal identity or ancestry (see target population). These census respondents make up the APS frame.

Sampling design and stratification:
An important part of stratification uses the survey's domains of estimation, which are groups of units for which estimates are targeted. These domains of estimation corresponded to geographical regions for which estimates with an "acceptable" level of precision for a particular Aboriginal group (i.e. First Nations, Métis or Inuit) and particular age group were targeted.
More precisely, the domains of estimation were created by cross-tabulating the following variables:

Geography
----Inuit regions
----Outside Inuit regions
--------province/territory
--------Atlantic provinces grouped

Age group
----18 to 24 years of age
----25 to 54 years of age
----55 or more years of age

Aboriginal group
----Inuit in Inuit regions
----Inuit outside Inuit regions (rest of Canada)
----Aboriginal groups (excluding Inuit) combined for Atlantic Canada (outside Nunatsiavut), Quebec (outside Nunavik), Yukon and NWT (outside Inuvialuit)
----For Ontario, Manitoba, Saskatchewan, Alberta and British Columbia
--------Status First Nations people living off reserve
--------Non-Status First Nations people living off reserve
--------Métis

Stratification will produce more precise estimates if units are homogeneous within strata and heterogeneous between strata. In addition, the estimation weights associated with survey respondents should ideally be as close as possible within strata.

Within each domain of estimation, three variables were used to create the 2017 APS strata:
- the region associated with the census form type (2A-L or 2A-R),
- whether the individual self-responded to the census or responded during non-response follow-up (NRFU), and
- the type of Aboriginal identification on the 2016 Census (Aboriginal identity or Aboriginal ancestry only).
The APS design can be considered a two-phase design in which the first phase corresponds to the selection of the 2016 Census long-form sample and the second phase corresponds to the selection of the APS sample.

Allocation method:
A method for optimal allocation between the substrata of a particular domain was used, taking into account different types of sample size loss, such as expected non-response and the probability of each unit belonging to the target population.

Sample size:
The final sample of the 2017 APS contained a total of 43,645 units.

For more details on the sampling design, the domains of estimation, the stratification and the allocation, please consult the Aboriginal Peoples Survey, 2017: Concepts and Methods Guide, which is available at the Related Products on the Integrated Metadatabase (IMDB) webpage.
http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=3250

Data sources

Data collection for this reference period: 2017-01-16 to 2017-08-15

Responding to this survey is voluntary.

Data are collected directly from survey respondents and derived from other Statistics Canada surveys.

Before the start of collection, introductory letters explaining the purpose of the survey were sent to the selected respondents.

The questions in the 2017 APS were administered in a computer assisted interviewing (CAI) environment. Two computer-assisted interview methods were used for this survey: Computer Assisted Telephone Interviews (CATI) and Computer Assisted Personal Interviews (CAPI). In 2017, CAPI was used for all Inuit regions, the Northwest Territories (excluding parts of Yellowknife) and in some parts of the Yukon. CATI was the primary mode of collection for dwellings in the provinces.

Respondents were interviewed in the official language of their choice. For Inuit regions, the questionnaire was translated as a paper copy into Inuktitut and Inuinnaqtun. On-screen help instructions were also available in the Inuit languages. These instructions could include: include/exclude statements, definitions, examples and/or supplementary instructions.

The time required to complete the survey varied from person to person. In some cases, the 2017 APS interview took up to an hour or more to finish, but on average the survey took about 40 minutes to complete.

Proxy reporting was acceptable in some circumstances (for example, when the respondent was unable to answer for health related reasons, due to a language barrier, or because the selected respondent was away from home for the duration of the survey.) Any member of the household over the age of 18 could act as a proxy for the selected respondent and answered the survey for them.

More than 43,000 individuals were selected to participate to the 2017 APS. Of those, approximately 32,330 respondents completed the 2017 APS for a response rate of 76%. Excluding 8,380 non-Aboriginal respondents, the total number of Aboriginal respondents included in the 2017 APS database is 24,220 (including the approximately 280 additional respondents from the APS - Nunavut Inuit Supplement - see estimation).

The 2017 APS sample was drawn from respondents who reported either Aboriginal identity or Aboriginal ancestry in the 2016 Census of Population long-form. APS respondents were told that Statistics Canada planned to combine their APS and census responses. Accordingly, the final edited Aboriginal Peoples Survey master microdata file was linked with the 2016 Census of Population Dissemination Database. In the end, more than 250 census variables were added to the final APS file for 2017.

The specific benefits of an APS-Census record linkage are reduced response burden for the target population of the APS, the derivation of survey weights which are crucial to providing valid estimates, and the creation of a comprehensive microdata file which can be used by data analysts to extend their learning and to inform policy and program development for Aboriginal peoples in Canada.

All products containing linked data are disseminated in accordance with Statistics Canada's policies, guidelines and standards. Only aggregate statistical estimates that conform to the confidentially provisions of the Statistics Act are released.

View the Questionnaire(s) and reporting guide(s) .

Error detection

Responses to the 2017 APS were captured directly by the interviewer at the time of the interview using a computerized questionnaire. In many cases when a particular response appeared to be inconsistent with previous answers or outside of expected values, the interviewer was prompted, through message screens on the computer, to confirm answers with the respondent, and, if needed, to modify the information directly at the time of interview. This editing, however, was conducted only with errors that were fairly simple and straightforward to detect and fix. These edits were applied at the micro level.

The collected data were then subjected to further editing processes in order to correct errors that required more complex edit rules. Customized edits consisted of validity checks within and across variables to identify gaps, inconsistencies, and other problems in the data, and corrections were performed based on logical edit rules. Editing at this stage was also applied at the micro level, using SAS (Statistical Analysis System).

Imputation

For the 2017 APS, a series of important imputations was conducted in relation to Aboriginal identity classifications. For example, those with missing data for questions ID_Q10 on Aboriginal identity group, ID_Q25 on Registered Indian Status, or ID_Q30 on membership in a First Nation or Indian band were imputed values based on their responses to the census. For those who self-reported as an Aboriginal person on APS question ID_Q05 but who did not report any specific Aboriginal group in ID_Q10, an imputation was also conducted based on the respondent's answer to the Census.

Estimation

The initial weight of a unit in a given APS stratum corresponds to the product of two components: the inverse of the stratum sampling fraction and the Census weight of the unit in question. The stratum sampling fraction is calculated as the number of people selected for the APS in each stratum divided by the total number of available Census long-form respondents for that stratum. The weights were then adjusted for non-response.

Two adjustments were made for two types of non-response: non-contact and non-response with contact (mainly refusals). First, a logistic regression model was constructed for each adjustment to predict the probabilities of being contacted or of responding when contacted on the basis of Census variables and collection variables known as "paradata" (number of contact attempts, for example). Second, respondents and non-respondents with similar predicted response probabilities were assigned to adjustment classes using cluster analysis. Third, the inverse of the weighted response rate in a class was used as the adjustment factor for that class, and the weights of the responding units within the class were adjusted accordingly.

Next, two post-stratification adjustments were made. The first post-stratification ensured that the sample did not under represent or over represent certain combinations of Aboriginal groups, regions and age groups of the Census. The second post-stratification ensured that the Aboriginal identity population estimated from the APS screening questions corresponded to the population defined from the Census screening questions within each post-stratum defined by the cross-tabulation of region, Aboriginal identity group and age group.

The Sigma-gap method was then used to detect and reduce excessively large weights within each post-stratum. After the weights were sorted in descending order, the excessively large weights were reduced to the value of the first non-outlier weight. The mass of the reduced weights was then redistributed proportionally within the post-strata.

Lastly, respondents from the APS - Nunavut Inuit Supplement sample that completed the APS questionnaire but were identified as out-of-scope for the Nunavut Inuit Supplement because they were not enrolled under the Nunavut Agreement were added to the APS sample with a weight of one. The APS sample units in Nunavut that were not enrolled under the Nunavut Agreement were then reweighted within the second post-stratification adjustment groups in order to maintain the previously achieved control totals. More details about the APS - Nunavut Inuit Supplement are available here:
http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=5270

For the 2017 APS, the bootstrap method was used to calculate the variance. For the sole purpose of calculating the variance, the 2016 Census was seen to have two phases: the initial sample of approximately 1 in 4 dwellings as the first phase and census respondents as the second phase. Although the final response rate was quite high for the 2016 Census (97.8% for the long form), this second phase ensures that the variance calculation takes into account the non-response that occurred. The two phases of the Census were later combined into a single phase. The APS sampling was treated as a second phase, and then the general bootstrap method for two-phase sampling developed for the 2006 APS was used (see Langlet, É., Beaumont, J.-F., and Lavallée, P. 2008. "Bootstrap Methods for Two-Phase Sampling Applicable to Postcensal Surveys". Paper submitted to Statistics Canada's Advisory Committee on Statistical Methods, May 2008, Ottawa).

For the APS, 1000 sets of bootstrap weights were generated using this method. The method can lead to negative bootstrap weights. To overcome this problem, a transformation was done on the bootstrap weights that reduced their variability. Therefore, the variance calculated on these transformed bootstrap weights has to be multiplied by a factor that is a function of a certain parameter, called phi. The value of the parameter corresponds to the smallest integer that makes all bootstrap weights positive. For the APS, this parameter has a value of 4. The variances calculated on the transformed bootstrap weights have to be multiplied by four squared, that is 16. In addition, the CVs obtained (square root of the variance divided by the estimate itself) have to be multiplied by 4.

Quality evaluation

Differences between the APS and other data sources:
Due to a number of differences in methodology between the 2017 APS, previous APS cycles and other Statistics Canada surveys, comparisons of data between sources should be done with caution.

2017 APS and 2016 Census:
The census and the APS are both rich sources of information on Aboriginal peoples that complement each other. The APS takes concepts that are touched on in the census and asks questions that dig deeper in order to provide more detailed information. For instance, the census provides information on labour market activities (which includes: labour force status, class of worker, industry, occupation and work activity during the reference year; from questions 30 to 49). Adding information from the APS provides an opportunity to learn more about part-time employment, permanent work, job satisfaction, looking for work, labour market attachment, past job attachment, labour mobility and other labour activities.

The APS also covers entire topics or themes that are not included in the census. For example, the APS can provide detailed information on education and health of Aboriginal peoples.

Although both surveys cover the "identity population" by design, the 2017 APS, like the 2012 APS, did not cover the "ancestry-only population". (Census respondents reporting Aboriginal ancestry-only were part of the APS sample because they had a non-negligible probability of reporting identity on the APS, and these respondents only remained in the APS data set if they actually reported Aboriginal identity in the APS.)

In general, the Aboriginal identity population counts on the 2017 APS for certain subpopulations may differ from those obtained from the census, even if the population universe for the census is restricted to that of the APS. The second post-stratification ensured that the number of individuals with Aboriginal identity was the same in the census and the APS, but this applied only to certain combinations of Aboriginal group, region and age group. However, the Aboriginal identity population counts may differ for other subpopulations which were not controlled for during post-stratification. Moreover, for a given individual, the Aboriginal identity reported may differ in some cases between the census and the APS. There are a number of reasons why Aboriginal identity may not be the same on both surveys. The differences could be the results of the following factors:

- Different interview methods
- Proxy effect
- Different questionnaires
- Different contexts
- Effect of time
- Processing

2017 APS and 2012 APS:
The most significant difference between the 2017 APS and the 2012 APS is the addition of the 'Aboriginal responses not included elsewhere' identity group (which includes individuals reporting being a Status Indian or member of a First Nation/Indian band only). In 2012, members of this group were imputed to being a First Nations person. Therefore, when comparing First Nations estimates between the 2012 and 2017 APS, 2017 APS respondents who were either Status Indian or a member of a First Nation/Indian band and who did not self-report as Aboriginal should be included in the estimate for First Nations people. It should be noted that although 'Aboriginal responses not included elsewhere' was kept as a distinct identity group in 2017, individuals in this group were combined with First Nations individuals during the post-stratification as in 2012.

Another important difference in methodology is the fact that the 2017 APS sample was selected from respondents to the 2016 Census, while the 2012 APS sample was selected from respondents to the 2011 National Household Survey (NHS). The characteristics of respondents to the census may be different than those of respondents to the NHS. The fact that non-respondents have different characteristics than respondents creates what is called non-response bias. Despite the fact that the NHS used follow-up strategies and non-response adjustment strategies at weighting to reduce this bias, it is possible that some non-response bias still remains.

2017 APS and APS-Nunavut Inuit Supplement:
There are many methodological differences between the 2017 APS and the APS-Nunavut Inuit Supplement (APS-NIS). To begin, the populations covered by each survey are not the same. The APS-NIS only includes data for Inuit enrolled under the Nunavut Agreement (NA) while the 2017 APS includes data for all Inuit (and all other Aboriginal identity groups).

Moreover, the domains of interest and sampling strata were not the same for the two surveys. The 2017 APS sample was selected based on domains of interest defined using geography (Inuit regions, province/territory, Atlantic provinces grouped), Aboriginal group and age group. In comparison, the APS-NIS sample was selected based on domains of interest defined by Nunavut community and education group. In fact, the APS-NIS was designed to produce community-level estimates in Nunavut whereas the 2017 APS was designed to produce estimates only at the Nunavut level.

The domains of interest of each survey also impacted the weighting strategy. For the 2017 APS, the post-stratification produced weights so that population counts by geography, Aboriginal group and age group matched 2016 Census totals. For the APS-NIS, the post-stratification was done for each education group within each Nunavut community. The difference in the weighting strategies can create differences between the estimates produced for the two surveys.

When comparing estimates of the 2017 APS content across different regions in Canada (e.g. comparing estimates for different Inuit regions), the 2017 APS sample should be used to produce the estimates.

Estimates for the APS-NIS questions should be produced using the APS-NIS sample. The 2017 APS content for this sample should be used as auxiliary information to enhance the analysis of Inuit enrolled under the NA.

More information about the APS-NIS can be found here:
http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=5270

2017 APS and Canadian Survey on Disability:
The Disability Screening Questions (DSQ) module, included for the first time in the 2017 APS, serves to identify persons with disabilities (PWD), thus allowing comparisons of measured survey characteristics between PWD and persons without disabilities (PWoD).

The official source for disability-specific data in Canada, such as prevalence and counts, is the Canadian Survey on Disability (CSD), a survey which takes place every five years following the census. The CSD produces estimates of overall disability prevalence, prevalence by type of disability and by level of severity.

However, one exception is made for the APS. Since the APS has a much larger sample of Aboriginal persons with a disability than the CSD and the APS sample is considered more representative of the Aboriginal population, the APS is the official source of disability rates for Aboriginal persons.

To ensure that the APS rates are comparable to those of the CSD, the CSD methodology was applied to the APS: anyone in the APS sample who did not report having difficulty or a health problem or long-term condition to the Activities of Daily Living question on the 2016 Census of Population was considered not to have a disability, regardless of their answers to the DSQ on the APS. Moreover, to ensure comparability between the APS and the CSD when calculating disability rates (for example to compare disability rates of Aboriginal persons to those of the entire population), the age of individuals as of Census Day (May 10, 2016) must be used to exclude any individual under the age of 15 as of Census Day. For more information on the CSD methodology, refer to the 2017 CSD Concepts and Methods Guide.

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects which 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.

Revisions and seasonal adjustment

This methodology type does not apply to this survey.

Data accuracy

Two types of errors occur in surveys: sampling errors and non-sampling errors.

Sampling Errors:
The sampling error measure used for the APS is the coefficient of variation (CV) of the estimate, which is the standard error of the estimate divided by the estimate itself. In this survey, when the CV of an estimate is less than or equal to 16.6%, the estimate can be used without restriction. When the CV is greater than 16.6% but less than or equal to 33.3%, the estimate will be accompanied by the letter "E" to indicate that the data should be used with caution. When the CV of an estimate is greater than 33.3%, or if an estimate is based on less than 10 units, the cell estimate will be replaced by the letter "F" to indicate that the estimate was suppressed for reliability reasons.

Non-sampling Errors:
Non-sampling errors arise primarily from the following sources: non-response, coverage, measurement and processing. The response rate for the APS was 76%. Total non-response will produce a bias if non-respondents have different characteristics from respondents and if non-response is not corrected properly. Non-response adjustments, combined with a relatively high response rate, helped reduce this risk of bias substantially. Non-response to specific questions is often due to difficulty understanding the questions. Thorough quality reviews and questionnaire testing were carried out before the survey, which reduced the extent of partial non-response. Cases in which there was a large proportion of missing responses to key questions were treated as a special form of total non-response.

Coverage errors occur when there are differences between the target population and the sampled population (or survey population). In particular, under-coverage can be problematic. Because the APS sample was selected from those who had participated in the 2016 Census, individuals who did not participate in the census could not be sampled for the APS. If this group of individuals is significantly different than the ones who participated in the census with respect to the characteristics measured in the APS, a bias could be introduced. This bias is assumed to be relatively small given the very high response rate obtained in the census (97.8% response rate for the long form).

Measurement errors occur when the response provided differs from the real value. Such errors may be attributable to the respondent, the interviewer, the questionnaire or the collection method, for example. For the 2017 APS, every effort was made to develop questions that would be understandable, relevant and appropriate for respondents. Other measures were also taken, including the use of skilled interviewers, extensive training of interviewers, and observation and monitoring of interviewers.

Processing errors may occur at various stages, including data capture, coding and editing. Quality control procedures were applied at every stage of data processing to reduce this type of error.

Date modified: