about national household survey data
Contact Peel Public Health to submit:
Need help understanding the data contained on this site?
- For each data source, information about data collection and analytical methods, limitations, the citation and additional resources can be found on the Data Sources and Methods page.
- Original source: Statistics Canada
- Distributed by: Statistics Canada
- Cite as: National Household Survey, 2011, Statistics Canada.
Data Collection Methods
- In June 2010, the Government of Canada announced the discontinuation of the mandatory long form census and the implementation of a voluntary household survey.
- The National Household Survey (NHS) was implemented for the first time between May and August 2011. The reference date for the survey is May 10, 2011 (the same day as Census Day).
- The target population for the NHS was all persons who usually live in Canada, in the provinces and territories, at the time of the survey. This includes those on Indian reserves or settlements, permanent residents and non-permanent residents (including refugees and work permit holders and their families).
- The NHS excludes the following groups:
- Those who live in collective dwellings (e.g., hospitals, nursing homes, penitentiaries)
- Canadian citizens living in other countries
- Full-time members of the Canadian Forces stationed outside of Canada
- Those living in non-institutional dwellings (e.g., work camps, hotels and motels, student residences)
- Foreign residents in Canada (e.g., representatives of foreign governments, members of armed forces of another country stationed in Canada).
- The NHS included 54 questions related to the individual and 10 questions about their dwelling.
- The NHS was conducted using two methods: self-administered survey (online or paper version) or enumerator-administered survey (Indian reserves and remote areas).
- The NHS is available in 31 languages other than English or French.
- The data consist of a random sample of 4.5 million dwellings from the 2011 Census of Population, representing approximately 30% of private dwellings in Canada.
- The national response rate was 68.6%. The response rate varied by province and census subdivision.
- Missing values for survey items were imputed by Statistics Canada using the “nearest neighbour” method.
- The sample was weighted to reflect the overall population in three steps: developing sample weights, adjusting for non-response and calibrating the weights against census totals.
- Prior to each data release, Statistics Canada certifies each estimate to determine whether the estimate will be released within the standard NHS products (based on a number of data quality factors).
- Data are obtained in Beyond 20/20*.ivt format and are extracted by the relevant geographies (province, census division, census subdivision).
- Due to the voluntary nature of the NHS, there is a possibility of non-response bias if those who chose not to respond to the survey are systematically different than those who did respond.
- Caution must be used when comparing NHS estimates to the 2006 long form census information because it is impossible to determine with certainty whether differences are due to actual changes in the population or due to non-response bias.
- Where common information is collected in the 2011 Census of Population and the 2011 NHS, the Census of Population should be considered to have higher quality and the Census estimates should be used over the NHS estimates.
- There were differences in the questions used to capture some concepts between the 2006 long form census and the 2011 NHS. Therefore, comparisons between the 2006 and 2011 variables should be made with caution.
- The NHS overestimated some population groups (e.g., population born in the Philippines, percent of population with a university certificate or diploma below a bachelors level) and underestimated other population groups (e.g., population born in Pakistan, recent immigrants).
- Trends in low income estimates from the NHS compared to estimates from previous censuses show markedly different trends than those derived from other surveys and administrative data sources. Therefore, low income data from the NHS should be used with caution and comparisons to previous census data should not be made.
- Not all Indian reserves or settlements were completely enumerated during the collection period of the 2011 NHS, including 13 reserves or settlements in Northern Ontario. The biggest impact of this incomplete enumeration will be seen for variables associated with Aboriginal or First Nation identity or language.
- Data from Statistics Canada are transformed using a random rounding process to maintain confidentiality. Values, including totals, are randomly rounded either up or down to a multiple of 5 or 10. The result is that, when these data are summed or grouped, the total value may not match the sum of the individual values, since the total and subtotals are independently rounded. Similarly, percentages calculated on rounded data may not necessarily add up to 100%. Note also that the same value in the same table may be rounded up in one analysis and rounded down in the next.
REFERENCES AND RESOURCES
- Statistics Canada – National Household Survey Profile
Last updated: February 12, 2019