The National Household Survey (NHS) was a supplement to the 2011 Census which was used only in 2011 to replace the long-form Census that year.
In 2011, all Canadian households completed the short Census. In previous Census years, and again in 2016, one in five households also completed the 'long-form Census' with almost 60 questions. This 20% sample was representative of the full Canadian population. In 2011, this mandatory 'long-form Census' was replaced with the voluntary NHS containing 64 questions. Most of these were identical to the previous long-form Census, but the survey method was different.
A large sample of Canadian households was asked to complete the NHS, but because it was not required by law, many households did not do so. Hence, the NHS results may not reflect the full Canadian population as accurately as the former long-form Census. Statistics Canada did not release data from areas with poor NHS response rates, and collapsed more detailed variables into broader categories in order to ensure data quality and maintain confidentiality. The result was a less detailed picture of Canadians than in previous Census years.
What we used to call "Census data", was actually two separate collections of data in 2011. Because of the different methods used, data from the 2011 Census and the 2011 NHS cannot be combined. Likewise, caution is advised in comparing data from the 2011 NHS and other Censuses.
Difference between NHS and Census
2011 Census:
Statistics Canada takes great care to ensure that no publicly-released Census data can be used to identify an individual person or business.
This means that some data cannot be released, or must be combined (aggregated) into larger categories or bigger geographic areas to prevent such disclosure.