Data based on the 2011 NHS, previous censuses, and
the Survey of Labour and Income Dynamics may not
be strictly comparable due to methodological differences.
Statistics Canada has advised caution in comparing
Census-based and NHS-based estimates. This is because
the change of methodology from a mandatory to a
voluntary survey introduces some uncertainty as to what
extent differences are due to actual changes in what is
being measured or to what Statistics Canada refers to
as survey non-response bias. Non-response bias is a
potential source of error for all surveys. It arises when the
characteristics of those who choose to participate in the
survey are different than those who refuse, and increases
as the response rate declines. Generally, the risk of
error increases for lower levels of geography and for
smaller populations. See 2011 National Household
Survey (NHS): Design and Quality, presentation to
the Housing Data Working Group
of the National
Housing Research Committee on November
While the change from a mandatory to voluntary survey
may have affected comparability of the data from the
NHS to earlier censuses, for 2011 the NHS is the most
comprehensive source of data on Canadian
The survey sampled about 4.5 million households, with
68.6% (about 3 million households) responding. The
NHS continues the Census’s tradition of providing
detailed data not available from other surveys in Canada;
useful data from the NHS are available at the municipal
and even neighbourhood levels (Statistics Canada has
suppressed data for some communities for confidentiality,
as well as data quality and non-response issues), as well
as for Canada, the provinces and territories. Further,
because of the large sample size of the survey, combinations
of many variables can be analyzed. When using the
NHS to estimate core housing
need, we undertook a
comprehensive review of the data, including the inputs
into core housing need. The
housing need estimates
reasonably represent housing conditions
A CMA must have a total population of at least
100,000, of which 50,000 or more must live in the core.
A CA must have a core population of at least 10,000.
Excluding the territories, about 83% of households
assessed for core housing need lived
in CMAs or CAs,
according to data from the 2011 NHS.
Since the SLID sample of about 34,000 households is
much smaller than the 2006 Census and 2011 NHS
samples, SLID-based estimates have less precision than
estimates based on census or NHS data. Thus differences
between SLID-based estimates, either from year to year
or between categories or geographic areas, may not be
statistically significant. Where possible, the significance
of differences between estimates has been assessed using
measures of precision of the estimates [coefficients of
variation (CVs—the coefficient of variation (CV) is the
standard error divided by the estimate; the smaller the
CV, the more accurate the estimate)] provided by Statistics
Canada. Letter grades indicating quality levels for
estimates are provided in some tables:
“A” indicates excellent data quality, with a CV
of less than 2%.
“B” indicates very good quality, with a CV between
2% and 3.9%.
“C” indicates good quality, with a CV between 4%
“D” indicates acceptable quality, with a CV between
8% and 15.9%.
“E” indicates that the estimate should be used
with caution since its CV is 16% or more.
“F” indicates that the estimate has been suppressed
due to unacceptable data quality—it either
has a CV of more than 33% or it is based
on 25 observations or fewer.
Income estimates from the 2011 SLID are for 2011, and
shelter cost estimates are as of the first quarter of 2012.