Coverage Technical Report, Census of Population, 2016
6. Dwelling Classification Survey (DCS)

6.1 Introduction

The Dwelling Classification Survey (DCS) measures dwelling classification errors in the census. The DCS also estimates the number of usual residents in occupied non-response dwellings. Census data are adjusted for occupied non-response dwellings.^{Note 1} The DCS estimates are used in census processing to specify how many persons are to be imputed during whole household imputation (WHI).

One of the potential sources of error in a census is the misclassification of a dwelling. When a household does not return a questionnaire, the enumerator has to determine whether the dwelling is occupied. Two types of dwelling classification errors can occur in this situation. First, an occupied dwelling can be misclassified as unoccupied. This dwelling classification error results in census population undercoverage since the dwelling occupants will not be included in the census database. Second, an unoccupied dwelling can be misclassified as occupied. When this error occurs, no questionnaire will be received from this dwelling, and the dwelling will be subject to non-response follow-up (NRFU).^{Note 2} If the NRFU fails to correct the dwelling’s classification to unoccupied, the dwelling will continue to be considered a non-response dwelling and will be subject to imputation. This dwelling classification error results in population overcoverage—persons are added to the census database when, in fact, no one lives at that dwelling. Estimates from the DCS are used to adjust census data for these two coverage errors.

Additionally, a third type of dwelling classification error that the DCS measures is the error incurred when marginal dwellings or dwellings under construction are misclassified as dwellings. Since these dwellings are classified as unoccupied—and because only occupied dwellings can be classified as non-response dwellings, which are subject to imputation—this error does not result in population overcoverage. However, this error does result in dwelling overcoverage. Census data are not adjusted for these dwellings, so census estimates of the housing stock include some degree of overcoverage.

6.2 Methodology

6.2.1 Stratification and sample selection

The DCS target population included all dwellings classified as either unoccupied or non-response dwellings, excluding dwellings in collection units (CUs) with collective dwelling, canvasser and reserve collection types.^{Note 3} Those areas were excluded because of cost and operational considerations.

The DCS sample size was set at 1,730 CUs. The sampling frame consisted of all self‑enumeration CUs (i.e., mail-out, list/leave and seasonal CUs, where respondents completed their own census questionnaire). Consequently, Nunavut had no in-scope CUs, and the DCS was not conducted there. The sample design was as follows: first, all in-scope CUs in Yukon (42 CUs) and the Northwest Territories (21 CUs) formed one stratum. All these CUs were selected for the DCS sample with certainty. All the CUs in Prince Edward Island formed a second stratum from which a simple random sample of 49 CUs was selected.

The remaining CUs were grouped into urban and rural strata. A CU was considered urban if it had initially been part of a census metropolitan area (CMA) or a census agglomeration (CA) that had 40,000 or more occupied dwellings. Further, all the CUs within a crew leader district (CLD) were considered urban if more than 50% of the CUs in the CLD were urban. All the remaining CUs formed the rural strata. Urban CUs were stratified by CMA and CA. A simple random sample of at least five CUs was selected within each stratum. From past census data, it was determined that five CUs was an appropriate workload for an interviewer. There were 984 urban CUs in the sample. To control field costs, CUs in close proximity to each other were chosen for the rural sample. This was done via a two-stage stratified random sampling design. In the first stage, CLDs were selected within each province. In the second stage, five CUs were selected from each of the selected CLDs. There were 746 rural CUs in the sample.

Dwelling subsampling within a sampled CU occurred when the number of unoccupied and non-response dwellings exceeded a maximum dwelling parameter, which was 50 in mail-out and list/leave CUs, and 200 in seasonal CUs. Subsampling of in-scope dwellings occurred in 325 CUs. Otherwise, all unoccupied dwellings and non-response dwellings in the sampled CUs formed the DCS dwelling sample. A total of 38,714 unoccupied and 9,994 non-response dwellings were sampled in 2016. Table 6.2.1 shows the distribution of the sample by province and territory.

6.2.2 Field interviews

A DCS questionnaire was used to verify the true occupancy status on Census Day of sampled dwellings in the sampled CUs that were classified as unoccupied on Census Day, or that were classified as occupied but for which no census form was returned. Occupancy status was verified in late June and early July 2016.

Timing for DCS field interviews were left to the discretion of each regional office. To determine occupancy status and to collect other information, enumerators were instructed to contact current occupants, neighbours, landlords, or any other person with knowledge about the dwelling. Up to three contact attempts were made for each dwelling. If the dwelling was found to have been occupied on Census Day, the number of occupants on Census Day was also obtained, along with occupants’ sex and age, where possible.

6.2.3 Processing and estimation

All completed questionnaires were sent to head office in Ottawa for processing.

The questionnaires were sent for data capture (key entry). Once data capture was completed, the questionnaires were subjected to an extensive set of consistency edits. The questionnaires that failed the edits were examined manually to resolve inconsistencies.

At this point in the processing, the unoccupied dwellings and the non-response dwellings in the sample were separated, and the dwellings’ classifications were confirmed against the final census listing. The questionnaires completed for each sampled CU were matched to the final census listing of unoccupied dwellings. If a match could not be found, the sampled dwelling was discarded and no further processing was required. Dwellings listed as unoccupied on the census list for which no DCS questionnaire was received were considered as total non-response to the DCS and proceeded to the next processing step. Similarly, the final census listing of all dwellings for which a census questionnaire was not received (i.e., the dwelling was classified as a non-response dwelling) was used to establish which of the DCS dwellings for which a DCS questionnaire was not received would be considered as total non-response to the DCS, before continuing on to the next processing step.

Total non-response to the DCS was addressed by a weighting adjustment, and item imputation was used for item non-response. The procedure was the same for unoccupied dwellings and non-response dwellings. When there was no information for a dwelling, the design weights of the respondents were adjusted to account for the design weight of the non-respondents. The adjustment was done separately by geographic post-stratum—that is, for each of the Montréal, Toronto and Vancouver CMAs, for the remaining urban areas in each province and territory, and for the rural areas for each province and territory. Item non‑response for occupancy status, number of usual residents, and dwelling type was addressed by imputation. Occupancy status was imputed first and then used in the imputation of the other variables. Design weights were then adjusted so that the sum of the adjusted weights for each geographic post-stratum equalled the number of unoccupied/non-response dwellings. The household size, collected on the DCS questionnaire when a dwelling was found occupied, was used to produce an estimate—by household size—of the occupancy rate of private dwellings classified by the census as unoccupied and non-response.

6.2.4 Census whole household imputation (WHI)

The whole household imputation (WHI) procedure aims to represent non-response private dwellings in the census and, as such, imputes for total non-response to the census. In geographic areas where the DCS was conducted (i.e., CUs with mail-out and list/leave collection types), the DCS estimates were used as input to the WHI algorithm to impute occupied private dwellings based on post-strata and household size distributions. First, within a DCS geographic post‑stratum, all the non-response dwellings were identified (this was done separately for unoccupied and non-response dwelling universes). Second, only for the non-response dwelling universe, any non-response dwelling for which field collection had obtained the number of usual residents was deemed to be occupied and was assigned the recorded household size. Finally, non-response dwellings were randomly selected and imputed as occupied. The selection was done so that the final number of non-response and unoccupied dwellings converted to occupied dwellings in the post-strata equalled the DCS estimate of occupied dwellings in the non-response and unoccupied dwelling universes. In geographic areas where the DCS was not conducted (i.e., CUs with collective dwelling, canvasser and reserve collection types), all private dwellings classified as non-response were imputed as occupied based on the household size distribution of census respondents. The WHI procedure resulted in all private dwellings being classified as either occupied or unoccupied in the census database (i.e., there were no longer any non-response dwellings).

Once a private unoccupied/non-response dwelling was imputed as occupied, a procedure—subject to the constraints of DCS estimates by post-stratum and household dwelling size—was used to impute the household dwelling size and other variables. Household size was determined by randomly selecting a dwelling from all dwellings that had completed a census questionnaire in the same CU. The complete record from this donor household was then assigned to the unoccupied/non-response dwelling that was imputed as occupied. If no donor was found, then only a household size was assigned.

More information on WHI can be found in the report by Farr (2018).

6.3 Estimates

Census data were adjusted using DCS estimates, via the WHI procedure, for non-response dwellings and for occupied dwellings that were misclassified as unoccupied. The estimates are given in sections 6.3.1.1 and 6.3.2. Census data were not adjusted for marginal dwellings or dwellings under construction that were misclassified as dwellings. Section 6.3.1.2 presents estimates of the number of marginal dwellings and dwellings under construction that were classified in error as dwellings and that were therefore erroneously included in the housing stock.

6.3.1 Unoccupied dwellings

6.3.1.1 Occupied dwellings misclassified as unoccupied

Table 6.3.1.1.1 gives the estimated number of dwellings classified as unoccupied that should have been classified as occupied, and the corresponding error rate for unoccupied dwellings by urban and rural area,^{Note 4} and by province and territory. For comparison, Table 6.3.1.1.2 gives the same estimates for the 2011 Census. Table 6.3.1.1.3 gives the estimated number of persons living in occupied dwellings that were misclassified as unoccupied. Table 6.3.1.1.4 shows the number of households and persons added to the initial 2016 Census counts to adjust for these misclassifications.

Table 6.3.1.1.1 shows that 15.0% of all dwellings classified as unoccupied in 2016 were actually occupied. This is a slight increase from 13.8% in 2011. Dwelling misclassification was more prevalent in urban areas (21.3%) than in rural areas (6.5%). Urban areas show an increase from 2011. Increases in the misclassification rates occurred for all provinces except New Brunswick, where it decreased.

Among the provinces and territories, the Northwest Territories had the highest misclassification rate (19.8%), followed by British Columbia (18.0%), Alberta (16.9%), Ontario (16.5%), Yukon (16.3%) and Quebec (14.3%). The rates for the other provinces ranged from 12.9% for Manitoba to 5.9% for Prince Edward Island.

Because of errors in the initial dwelling classification, approximately 178,219 households were not enumerated in the 2016 Census. This is the number of households added to the census during WHI. Table 6.3.1.1.4 shows the number of households and persons added to adjust for occupied private dwellings misclassified as unoccupied.

6.3.1.2 Housing stock overcoverage

Table 6.3.1.2 shows the estimated number of unoccupied dwellings not in the housing stock, and the corresponding error rate for unoccupied dwellings for various geographic areas. No adjustments were made to the census database to account for dwellings not in the housing stock that were misclassified as unoccupied.

The enumeration of unoccupied dwellings that fall outside the housing universe results in dwelling overcoverage. Dwellings are considered to be outside the housing universe if they are used for commercial purposes, if they are not habitable year round, or if they are double counted in the census. Double counting can occur when the dwelling appears to have two addresses associated with it, or when two questionnaires are mistakenly returned for a dwelling that no longer contains a separate apartment.

The DCS estimates of the number of unoccupied dwellings misclassified as dwellings were not used to adjust the census database because of the degree of subjectivity associated with classifying a dwelling as suitable for year-round occupancy. A dwelling must have a source of heat or power and provide complete shelter from the elements to be considered suitable for year-round occupancy. It is sometimes difficult to tell whether or not a dwelling is habitable, such as when a dwelling is a cottage, is under construction and almost complete, or has deteriorated.

Dwellings outside the housing stock accounted for 9.8% of all dwellings classified as unoccupied. Among the provinces and territories, the incidence of dwellings outside the housing stock being classified as unoccupied ranged from 5.2% in Prince Edward Island to 17.2% in Yukon. The problem had a similar prevalence in urban areas (9.7%) and rural areas (10.0%).

Dwellings actually outside the housing stock represented 0.8% of all private dwellings in the 2016 Census. This was a slight increase from the 2011 error rate of 0.6%. Among the provinces and territories, the error ranged from a low of 0.1% in the Northwest Territories to a high of 1.4% in New Brunswick.

6.3.2 Non-response dwellings

6.3.2.1 Persons added in non-response dwellings

Table 6.3.2.1.1 gives the estimated number and rate of occupied non-response dwellings in the census by urban (population over 50,000) and rural area, and by province and territory. Table 6.3.2.1.2 gives the number of persons estimated by the DCS to be living in these non-response dwellings, and Table 6.3.2.1.3 gives the same information for the 2011 DCS.

Table 6.3.2.1.1 shows that 63.1% of all dwellings classified as non-response were actually occupied. The census did relatively the same job of classifying non-response dwellings in urban areas (63.5%) as it did in rural areas (61.4%). At the province and territory level in 2016, Quebec had the highest rate of correctly classified non-response dwellings, at 67.8%, while New Brunswick had the lowest rate, at 53.4%.

Table 6.3.2.1.2 shows the number of non-response dwellings in the 2016 Census, and gives the number of persons added in those dwellings through the DCS. Table 6.3.2.1.3 shows the same data from the 2011 DCS. In 2016, a total of 357,666 persons were added to the census in 179,823 dwellings. The comparable 2011 numbers were 443,098 persons in 220,181 dwellings.

6.3.2.2 Dwellings not in the housing stock misclassified as non-response

Table 6.3.2.2 shows the 2016 Census dwelling classification error from dwellings misclassified as non-response because they should not have been included in the housing stock. Section 6.3.1.2 provides the definition of dwellings outside the housing universe and comments on the difficulty in determining whether a dwelling should be included in the housing stock.

At the national level, dwellings outside the housing stock accounted for 4.0% of all non-response dwellings. The error rate was higher in rural areas (5.2%) than urban areas (3.7%). For provinces and territories, the incidence of dwellings outside the housing stock being classified as non-response ranged from 1.6% in Prince Edward Island to 9.2% in Manitoba. At the national level, non-response dwellings outside the housing stock accounted for 0.1% of all private dwellings. This error was rounded to 0.0% in Prince Edward Island, 0.2% in Manitoba and Yukon, and 0.1% in all other provinces and in the Northwest Territories.

Notes

• Previous
• Next