Respondent Selection Study for the General Social Survey

Detailed information for 2011




One Time

Record number:


This study will be used to determine which method would be the most effective to select households in Canada for any given survey that is conducted by Statistics Canada.

Data release - All collected data will only be used internally to determine the best sampling frame for the General Social Survey (GSS).


The main objectives of this study are to:
- evaluate the quality of the Address Register (AR) in a Computer Assisted Telephone Interview (CATI) environment,
- obtain specific information on household's telephone services,
- obtain phone contact information through mail,
- verify if tracing methods may improve the quality of the sample frame,
- evaluate if a dual base should be considered as a sampling frame.

The information collected through the study will include: the name, age and sex of each household member. The study will also verify and update the coordinates of each selected household. Additional questions about each of the household's landline and cellular phone numbers, and the education, dwelling characteristics, income and willingness of the respondent to do an internet survey questions, will also be asked.


  • Collection and questionnaires
  • Statistical methods

Data sources and methodology

Target population

Canadian population aged 15 and over for which the selected phone number is used as a home phone number. Businesses will be considered out of scope.

Instrument design

The content questionnaire was developed with the input of methodology, the Collection Systems and Infrastructure Division, the Collection Planning and Management Division team members based and/or adapted on existing Computer Assisted Telephone Interview (CATI) modules including harmonized content modules.

The CATI questionnaire comprises Standard Random Digit Dialing Entry/Exit modules and some specific content developed to collect each of the household's telephone numbers, including cellular phone numbers, as well as education, dwelling type, household income and willingness of the respondent to complete survey on the internet.

A paper questionnaire was also developed to be sent to addresses that cannot be reached by telephone. It consists of collecting rostering information for the household, the telephone number(s), including cellular phone numbers, the education level of all members, type of dwelling and income.

Both content questionnaire specifications and the paper questionnaire were reviewed by the Questionnaire Design Resource Centre. The proposed changes were adopted.


This is a sample survey with a cross-sectional design.

A sample of 4800 households equally distributed in the provinces of Quebec and Saskatchewan is used for this test. This sample is not representative of all households in the above provinces but rather representative of a combination of information available on the Address Register. The following stratification has been used:

1) Mailable civic address with at least one landline phone number

2) Mailable civic address with only cellular phone number(s)

3) Non mailable address with at least one landline phone number

4) Non mailable address with only cellular phone number(s)

5) Mailable civic address without a phone number

6) Non mailable address without a phone number

7) Phone number not found on the Address Register

Evaluate the quality of the Address Register (AR) in a Computer Assisted Telephone Interview (CATI) context:

The majority of units from the AR belong to the first four strata. Confirming the civic address will allow us to assess the quality of the link between the address and the phone number found on the AR. A few socio demographic questions will also be asked in order to compare those results with the information found on the AR.

Obtain specific information on household's telephone services:

Every household will be asked to provide a list of all phone numbers (landline and/or cellular) used to reach all members of the household as well as the name of the service providers. This information will be of utmost importance because it will be a determining factor on the choice of the future sample frame. Among other things, this information will be used in the weighting process.

Obtain phone contact information through mail:

For units coming from stratum (5), a letter will be sent to the selected addresses either to obtain one or several phone numbers through returning mail or to provide the household with a 1-800 phone number to contact us with. From this test, we hope to find out the expected response rate.

Verify if tracing methods may improve the quality of the sample frame:

The units from stratum (6) are, by definition, impossible to contact either through mail or by phone. Those units will be processed beforehand through tracing in order to establish a proportion of cases which, after the process, have been successfully linked to either a mailable address or a phone number.

Evaluate if a dual base should be considered:

In the event that a coverage problem arise from the use of only the Address Register (AR) (in a computer assisted telephone interview context), a telephone base with phone numbers not linked to the AR could possibly be used. Once we contact those phone numbers, we will try to explain the reason why they were not part of the AR.

Data sources

Data collection for this reference period: 2011-11-28 to 2011-12-16

Responding to this survey is voluntary.

Data are collected directly from survey respondents.

The data will be collected using a Computer Assisted Telephone Interview (CATI) application. Different introduction letters will be sent in advance to all respondents depending on their stratification. At the end of the two week CATI collection, a paper questionnaire will be sent to the households for which addresses were found in the Address Register, but were not confirmed as current address during interview time.

View the Questionnaire(s) and reporting guide(s) .

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects which could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.

No public use microdata file will be produced by Statistics Canada and data will not be made available through the Data Liberation Initiative (DLI).


Report a problem on this page

Is something not working? Is there information outdated? Can't find what you're looking for?

Please contact us and let us know how we can help you.

Privacy notice

Date modified: