Statistics Canada
Symbol of the Government of Canada
Warning View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

1. Estimates of population coverage error

1.1 Introduction

1.2 Net undercoverage

1.3 Undercoverage

1.4 Overcoverage

1.1 Introduction

The census defines the population to be counted and the rules by which the population is to be counted (see Section 4). Coverage errors occur when errors are made relative to these definitions and rules. Important sources of coverage error include the failure to include a dwelling and hence missing the people living there, respondent error by not including all persons who should be included or by including persons who should not be included. This section presents estimates of 2006 Census population net undercoverage, undercoverage and overcoverage. Both undercoverage and overcoverage may result in bias for census counts and estimates because the characteristics of those not included may differ from those who are included, and the characteristics of the duplicates may differ from those who were included only once. Net undercoverage states the extent to which the number of persons included in census data is higher or lower than complete enumeration.

1.2 Net undercoverage

The 2006 Census population net undercoverage rate was estimated at 2.67%1. This means that, on a net basis, the census missed 2.67% of those persons who should have been enumerated. The population undercoverage rate was estimated at 4.26% (1,384,372 persons) while the population overcoverage rate was estimated at 1.59% (515,715 persons). An undercoverage rate of 4.26% indicates that persons not included represent 4.26% of the census target population. An overcoverage rate of 1.59% indicates that duplicate enumerations represent 1.59% of the census target population.

Compared to the 2001 Census, coverage error has increased. The rate of undercoverage and the rate of overcoverage increased. Since the overcoverage rate increased more than the undercoverage rate, the estimated net undercoverage rate decreased.

This section presents estimates of net undercoverage for a variety of geographic and demographic variables: 

Table 1.2.2 gives the estimated net undercoverage in terms of the estimated net number of persons missed, the standard error of the estimate, and the corresponding estimated net undercoverage rate along with its standard error for each of these variables. Negative estimates of net undercoverage indicate that overcoverage was larger than undercoverage. See Section 9 to understand how this can occur.

The standard error provides a measure of the accuracy of the estimates resulting from sampling. The estimates are considered accurate to within plus or minus two standard errors 19 times out of 20. This means, for example, that there are approximately 19 chances in 20 (95%) that the real 2006 Census population net undercoverage rate falls within the range 2.34% to 3.00% (2.67% + or - two standard errors). Or, there are approximately two chances in three (68%) that the real rate falls within the range 2.50% to 2.84% (2.67% + or - one standard error).

Since net undercoverage is a reflection of both undercoverage and overcoverage, the reader is encouraged to also consult the estimates of undercoverage and overcoverage presented in Table 1.3. A low rate of net undercoverage, for example, may reflect low undercoverage or high undercoverage along with high overcoverage.  

Population net undercoverage was highest in Ontario and the western provinces. Among the provinces, the population net undercoverage rate was highest in Ontario at 3.69%, followed by Alberta at 3.27% and Manitoba at 2.90%. Quebec had the lowest provincial rate of population net undercoverage, 0.80%. Net undercoverage was higher in the territories, from 3.76% for the Northwest Territories to 5.61% for the Yukon Territory.

Population net undercoverage was usually higher for men, and highest for young adults. The rate of net undercoverage for males was slightly more than two and a half times the rate for women, 3.89% versus 1.48%. Net undercoverage was highest for those aged 20 to 34 for both males and females. Males aged 25 to 34 had the highest net undercoverage rate at 9.91% versus 9.46% for the younger 20 to 24 group. Among women, the rate was 5.69% for the 20 to 24 group and 6.08% for those aged 25 to 34. Net undercoverage was negative, indicating more persons counted more than once than considered as undercoverage, for males and females aged 15 to 17, for older women (55+) and for older men (65+).

Population net undercoverage for the 15+ population was higher for single persons. Considering marital status, about two thirds of net undercoverage for the 15+ population was from persons who had never been married and were not in a common-law relationship. The rate of net undercoverage for this group was 6.70%. Net undercoverage for persons who were separated and not in a common-law relationship was high, 9.75%, especially for males, 16.84%.

Population net undercoverage was higher for allophones. The net undercoverage rate for those whose mother tongue is English was larger than for those who reported French, 2.31% versus 0.52%. This reflects lower net undercoverage for Quebec. The rate of net undercoverage for allophones, those whose mother tongue is neither of the official languages, was higher, 6.89%. Net undercoverage for allophones approached the level of net undercoverage for those whose mother tongue is English, 385,432 persons versus 439,185 persons.

Population net undercoverage was slightly more common in urban areas. Among those who should have been enumerated in any of Canada's census metropolitan areas (CMAs), net undercoverage was 2.87%. This is slightly higher than the net undercoverage rate of 2.25% for those not living in CMAs.

1.3 Undercoverage

Persons counted as undercoverage are likely to not have been included as a usual resident on the questionnaire that was completed for their usual residence. Persons, for example, who consider their residence as temporary may not have been included as a usual resident elsewhere. Persons without a usual residence, who were otherwise not enumerated, e.g., the 'homeless population,' are also part of undercoverage.

This section presents estimates of undercoverage for a variety of geographic and demographic variables:

Table 1.3 gives the estimated undercoverage as the number of persons missed, the standard error of the estimate, the corresponding estimated undercoverage rate, and its standard error.
There are some occurrences of negative estimates of undercoverage such as -4,127 persons for women aged 55 to 64. See Section 9 to understand how this can occur.

There are some demographic trends in undercoverage:

  • The rate of undercoverage for men was almost twice the rate for women, 5.51% versus 3.04%.
  • For both men and women, undercoverage was highest for young adults aged 18 to 34.
  • Among young adult males, undercoverage was 10.06% for the 18 to 19 age group, 12.21% for the 20 to 24 age group and 11.42% for the 25 to 34 age group.
  • Considering marital status, undercoverage was highest in the population aged 15 or more for those who were not married nor in a common-law relationship. The rate for this group was 8.86%. Undercoverage was also high for separated persons who were not common-law. The rate for this group was 9.75%. In both cases, rates were higher for men than women.

A profile of the person most likely to have been missed in the 2006 Census emerges from Table 1.3 as male, between 18 and 34 years of age, and single (never married nor common‑law). Mother tongue other than English or French is also important.

1.4 Overcoverage

Population overcoverage is the number of enumerations in excess of persons who are included in census tabulations more than once, usually twice. This is an error resulting in bias for census counts and estimates because they should only have been included once. Examples of overcoverage include children whose parents have separate residences and each parent includes the children on their census form, persons who need to reside away from their family for reasons of work who are listed on their family's form and also on the form for the dwelling they live in while working, and students away at school who are listed both by their roommates and their parents.
 
This section presents estimates of overcoverage for a variety of geographic and demographic variables:

Table 1.3 also gives the estimated overcoverage in terms of the number of persons included more than once, the standard error of the estimate, the corresponding estimated overcoverage rate and its standard error.

The estimates of overcoverage in Table 1.3 are subject to less sampling error than the estimates of undercoverage. There are some demographic trends in overcoverage:

  • Across the provinces and territories, overcoverage varied much less than undercoverage did.
  • Overcoverage was only slightly higher for females than males, 1.62% versus 1.56%.
  • Overcoverage is highest for children and young adults from age 5 to 34. Rates are highest for young adults aged 20 to 24 at 2.88%.
  • As for undercoverage, overcoverage was highest neither for persons who were never married nor in a common-law relationship at 1.96%.

A profile of persons most likely to be counted twice emerges from Table 1.3 as equally likely to be male as female, equally likely to be a child or a young adult, and, if 15+, single or widowed. Analysis of overcoverage revealed that about half of the persons counted more than once were children with parents in separate households, or young adults away from their family home, or families who moved.

Note:

  1. This is different from the rate of 2.8% released on September 29, 2008 because incompletely enumerated Indian reserves and settlements are excluded. All of the estimates of coverage error in this document exclude coverage error for this group.

   Previous page | Table of contents | Next page