1. Data collection

1.1 Census data collection

The data collection stage of the 2011 Census ensured that each of the 14.5 million dwellings in Canada received a census questionnaire. The census enumerated the entire population of Canada, which consists of Canadian citizens (by birth and by naturalization), landed immigrants, and non-permanent residents together with family members living with them. Non-permanent residents are persons living in Canada who have a work or study permit or are claiming refugee status, and family members living with them.

The census also counted Canadian citizens and landed immigrants who were temporarily outside the country on Census Day. This included federal and provincial/territorial government employees working outside Canada, Canadian embassy staff posted to other countries, members of the Canadian Forces stationed abroad and Canadians aboard merchant vessels.

In the Census of Canada, various questionnaires and forms are used to collect data. Form 1 is the Visitation Record (VR). It is used to list every occupied and unoccupied private dwelling or collective dwelling, agricultural operation and agricultural operator in the collection unit. It serves as an address listing for field operations and control purposes for data collection.

Form 2A is the basic short questionnaire used to enumerate all private dwellings. Every household that receives a 2A census questionnaire is asked to list all household members who belong to the census population and answer questions for them.

Form 2C is used to enumerate Canadians posted in other countries, including government employees (federal and provincial/territorial) and their families, and members of the Canadian Forces and their families.

Form 3A is an individual census questionnaire used to enumerate persons in collective dwellings^Footnote1 (each person in a collective dwelling must complete a separate Form 3). It can also be used to enumerate usual residents in a private household who prefer to complete their own census questionnaire rather than be included in a 2A questionnaire.

Wave methodology is a new approach to data collection first used in the 2011 Census. Households are contacted at appropriate times to remind them to participate in the census and persuade them to complete the questionnaire. In each wave, households are provided with the information they need to respond. Based on the fact that every Canadian household is required by law to answer the census questions, the method is designed to encourage people to respond promptly online, while mitigating the risk of a decline in overall response and the need for costly field follow-up.

This new methodology varies with the collection method used to distribute the census materials for a given region. These collection methods are described in the next section. In 2011, Canadian households had the option of responding online, on a paper questionnaire (mail-back) or by contacting the Census Help Line.

1.1.1 Collection methods: Delivery of census questionnaires

The three questionnaire delivery methods used in the Canadian census are mail-out, list/leave and interview. To make census collection as efficient as possible, Canada is divided into small geographic units known as collection units (CUs). In the 2011 Census, there were some 46,000 CUs in Canada.

1.1.1.1 Mail-out

For mail-out CUs, the postal system is used to deliver the census materials. This method ensures effective, coordinated distribution, without the need to recruit and train a large contingent of enumerators. Mail-out CUs are typically in urban areas. While mail-out CUs now include about 80% of Canadian dwellings, they cover only a tiny fraction of the country's land area.

1.1.1.2 List/leave

List/leave CUs are typically in rural areas. In those areas, enumerators prepare a list of dwellings and deliver the census materials. About 18% of Canadian dwellings are in list/leave CUs, which cover a large portion of the country's land area.

1.1.1.3 Interview

Interview CUs are usually in remote or difficult to access places and in Aboriginal communities. To limit the number of trips that enumerators have to make to those places for follow-ups — trips that are often expensive and logistically complicated — they do more than prepare dwelling lists and deliver census materials; they also complete a questionnaire with each household on the spot. Interview CUs cover just over half of Canada's land area, but only about 2% of its dwellings.

1.1.2 Wave methodology in the census

Wave methodology was designed to encourage online response while offering an alternative for households that do not wish to complete their questionnaire on the Web. This approach has many advantages for response rates, questionnaire registration, question flow and data capture.

Wave methodology is applied differently in different CU groups. Three main groups of CUs were defined, and a different wave methodology was developed for each one. Because of the nature of interview CUs, however, no wave methodology was developed for them. The sections that follow and Figure 1.1.2.1 provide an overview of the wave methodology used in the 2011 Census.

1.1.2.1 Mail-out collection units – Wave 1 letter

First, a set of mail-out CU was identified so that the households most likely to respond online could be targeted and those least likely to respond to the census could be avoided. That set of CUs covers about 75% of the dwellings in mail-out areas. For that group of CUs, Wave 1 involved sending out only one letter asking households to complete the questionnaire online using the secure access code (SAC) provided or call an automated system on a toll-free line to have a paper questionnaire mailed to them. The Wave 1 letters were delivered by the postal system one week before Census Day (i.e., on May 3, 2011).

Wave 2 consisted of a reminder letter sent to all Wave 1 non-respondent households. The letter reminded the households that they were required by law to complete the census. Like the Wave 1 letter, it also provided the SAC and the toll-free telephone number. It was delivered to households between May 16 and 18, i.e., as early as six days after Census Day.

In Wave 3, a paper questionnaire was sent to non-respondent households. It was delivered to them between May 25 and May 31, i.e., as early as 15 days after Census Day. The households could still respond online using a SAC printed on the front cover of the questionnaire. The questionnaire was accompanied by a letter indicating that if the questionnaire was not completed by May 31, 2011, an enumerator would contact the household by telephone or in person to complete the questionnaire. It was also noted in the letter that if the household refused to answer the census questions, the case could be referred to the Public Prosecution Service of Canada, which would take appropriate action under the Statistics Act.

Wave 4, which began on June 1, 2011, consisted of field non-response follow-up (NRFU) and an automated reminder call. NRFU is described in Section 1.1.3 of this document.

1.1.2.2 Mail-out collection units – Wave 1 questionnaire

The second group of CU on which a variant of wave methodology was used is the set of other mail-out CUs. That set of CUs covers about 25% of the dwellings in mail-out areas. Households in those CUs were considered less likely to respond after receiving only a letter. Wave 1 for those CUs was the mailing of a paper questionnaire. The questionnaire provided a SAC, so that the household had the option of responding online. As in the case of the first group, Wave 1 took place one week before Census Day (i.e., on May 3, 2011). Wave 2 for the second group was the same as for the first group. In Wave 3, the group's non-respondent households for which a telephone number was available in the census frame received an automated reminder call on May 24. Wave 4 was the same as the first group.

1.1.2.3 List/leave collection units

The third group of CU on which another variant of wave methodology was used is the set of all list/leave CUs. In Wave 1, enumerators delivered a paper questionnaire to all dwellings in those CUs on or about May 3. The questionnaire also provided a SAC, so that the household had the option of responding online. In Wave 2, all dwellings in those CUs received a thank you / reminder card in the mail on May 10, whether they had responded or not, because it was generally impossible in those areas to send mail to civic addresses. Wave 3, the last wave, involved going directly to field non-response follow-up as of May 20.

1.1.3 Non-response follow-up

As mentioned in the previous section, the final wave in the wave methodology was non-response follow-up (NRFU). In that wave, enumerators telephoned and visited households that had not responded. Each non-respondent household for which a telephone number was available received an automated reminder call at the beginning of the NRFU period. The message reminded non-respondents of their legal obligation to respond to the census.

The enumerators had information from the Field Management System (FMS) to help them manage their work. A computerized system accessible over the Internet, the FMS was developed for the 2011 Census to facilitate the gathering of collection progress information.

1.1.4 Verification of dwellings' occupancy status

Before NRFU, field operations were also carried out for the dwelling occupancy verification (DOV). The purpose of DOV, which began on May 13, 2011, was to identify a significant number of dwellings that were unoccupied on Census Day or cancelled (addresses that are not private or collective dwellings) before NRFU started. Identifying such dwellings close to Census Day should make occupancy classification more accurate and perhaps easier to perform. DOV also reduces the NRFU workload, since any unoccupied or cancelled dwellings it identifies do not require follow-up.

Nevertheless, errors in classifying a dwelling as occupied or unoccupied do occur during DOV and NRFU. Some dwellings classified as unoccupied are in fact occupied, and some non-respondent dwellings are unoccupied. As a result, another operation, the Dwelling Classification Survey, is carried out after NRFU. It assesses and determines the occupancy status of dwellings for which no completed questionnaire has been received (unoccupied dwellings, non-respondent dwellings or unresolved cases). The survey's results are used to adjust the Census of Population counts during processing (see Section 2.7 or the Coverage Technical Report, 2011 Census, Catalogue no 98-303-X).

1.2 Collection methodology of the 2011 National Household Survey

The target population of the National Household Survey (NHS) consists of persons enumerated by the census who live in private dwellings.

Two questionnaires are used to collect the data: Forms N1 and N2. The two forms contain the same questions, but Form N2 is used in interview areas. The same 10 census questions are also on the NHS forms.

The decision was made to conduct the NHS during the same period as the census in order to take advantage of census resources and infrastructure such as collection management systems and employees. However, that strategy had the potential to impose a heavier burden on employees and respondents. NHS collection operations therefore had to be managed in such a way as to minimize any impact on the results of the census, since the latter takes precedence. Persons who responded to the census on the Internet (mail-out CUs) were given the opportunity to complete the NHS online immediately after finishing the census questionnaire, and those who used a paper census questionnaire were contacted about completing the NHS on or after June 7, when the paper NHS questionnaires were delivered.

The NHS collection strategy also had to take account of resource limitations, time constraints and the survey's voluntary nature, which prevented interviewers from pushing too hard and reduced the number of possible contacts with non-respondents.

1.2.1 NHS collection in interview areas

In interview areas, NHS data collection with Form N2 was carried out at the same time as census enumeration. The responses to the census questions, which were also in Form N2, were copied from one form to the other, as long as the respondent agreed to participate in the NHS.

In non-interview areas, the wave collection methodology described in the next section was used.

1.2.2 Wave methodology in the NHS

A wave methodology was also used for NHS collection, in conjunction with the wave methodology used in the census. The NHS wave methodology was based not only on the CU's collection method but also on the census response status and response mode (online or paper). With the dwellings selected for the NHS, the three groups described below were established.

The first group of dwellings, referred to as Survey group 1, consisted of online census respondents. The dwellings in Survey group 2 were census respondents who used the paper questionnaire. Survey group 3 consisted of census non-respondents.

The NHS wave methodology is different for each survey group; details are presented below. Table 1.2.2.1 also provides an overview. As in the case of the census, because of the nature of the interview CUs, wave methodology was not used for them.

Table 1.2.2.1
Overview of the wave methodology in the 2011 NHS

1.2.2.1 Mail-out – Survey group 1

Online census respondents selected in the NHS sample were invited to complete the NHS questionnaire immediately after finishing the census. After submitting their online questionnaire (using the original secure access code), these respondents were automatically routed to a transition page informing them that they had been selected for the NHS. The page also contained basic information about the survey, including an explanation of its importance and a note concerning its voluntary nature. Respondents who wanted to continue were then redirected to the NHS questionnaire, and the application automatically displayed the first few questions, which were the same as the census questions that the respondents had just answered. This was Wave 1 for Survey group 1.

Wave 2 for Survey group 1 was the mailing of a reminder letter to non-respondent households. The letter, delivered to households on or about June 7, made reference to the NHS's importance.

No further materials were sent out after Wave 2 (i.e., no Wave 3), and the next stage was field non-response follow-up (NRFU). Starting on June 8, follow-up was phased in, CU by CU, depending on the level of response to the census and the possibility of redeploying enumerators to help out with census collection in areas that were behind schedule.

Field procedures for the conversion of NHS non-response were similar to the census procedures. Census employees, infrastructures and systems were also used. When operationally feasible, attempts to convert refusals were made in person by team leaders and enumerators who had received special training. The definition of a refusal was the same as in the census: a clearly expressed refusal to respond to the survey. However, in view of the NHS's voluntary nature, Statistics Canada's usual refusal rules for voluntary surveys were applied: i.e., termination of follow-up efforts after two firm refusals (including telephone refusals).

In addition, some interviewers from Statistics Canada's computer-assisted telephone interviewing centres were used to supplement NHS NRFU. Those interviewers were deployed mostly in areas where NHS collection was behind schedule and there were not enough field employees.

It is worth noting that dwellings in Survey group 1 never received a paper NHS questionnaire. Since those dwellings chose to respond to the census online, it was assumed that they would do likewise for the NHS if they agreed to respond. Mailing out a paper questionnaire would have had little impact on the response rate and would have increased collection costs significantly.

1.2.2.2 Mail-out – Survey group 2

There was no indication in the paper census questionnaire that the dwelling might receive an additional, voluntary survey later. As a result, dwellings that used a paper census questionnaire in May did not have their initial contact with the NHS until about four weeks after Census Day (on or about June 7), when they received a paper NHS questionnaire in the mail. A secure access code for online response was printed on each questionnaire. This was Wave 1 for that group.

For Wave 2, a reminder letter was delivered to non-respondent households one week later, on or about June 14. Like the reminder letter sent to Survey group 1, this letter reminded non-respondents of the importance of completing the voluntary survey.

In the original plans, Wave 3 involved mailing a second paper questionnaire to non-respondent households. That questionnaire, accompanied by another reminder letter, was to have been delivered on or about July 6. However, because of the uncertainty generated by the possibility of a work stoppage at Canada Post and the opportunity to start field follow-up earlier, Wave 3 was cancelled.

Field NRFU (Wave 4) began when census collection was sufficiently advanced. The same procedures as those described above for Survey group 1 were followed.

1.2.2.3 Mail-out – Survey group 3

The wave methodology used for census non-respondents was simpler than the methodology used for the other survey groups. Since the respondents of dwellings in Survey group 3 did not react positively to any of the requests to respond to the census, which is compulsory, sending those letters or questionnaires or leaving additional voice messages urging them to complete a voluntary survey would not have had a significant effect. The plan was to send out only the NHS questionnaire for delivery on or about June 7, but that wave was also cancelled because of the possible work stoppage at Canada Post, so that priority could be given to preparing and distributing Survey group 2 questionnaires, which was taking place at the same time.

Hence, the only contact with Survey group 3 households was Wave 4, NRFU. That wave, which started on June 8, was carried out during the same field visit as census follow-up.

1.2.2.4 List/leave – Survey groups 1, 2 and 3

The dwellings in list/leave CUs received their census questionnaires in early May. As in the case of the wave methodology used for the mail-out CU survey groups, dwellings that responded to the census online using the secure access code printed on the questionnaire were invited to complete the NHS questionnaire online immediately after finishing the census questionnaire. In addition, NHS questionnaires were delivered by enumerators on or about June 7 to dwellings that used a paper census questionnaire or did not respond to the census, and to dwellings that refused the NHS offer after completing the census questionnaire online.

No other materials were delivered to the door, and telephone or field NRFU started on the same dates as for mail-out areas. For Survey group 3, NRFU for the NHS was also carried out during the same field visit as NRFU for the census.