In order to carry out sampling for Statistics Canada's 2015 General Social Survey (GSS), each of the ten provinces was divided into strata (i.e. - geographic areas). Many of the Census Metropolitan Areas (CMAs) were each considered separate strata. This was the case for St. John's, Halifax, Saint John, Montreal, Quebec City, Toronto, Ottawa, Hamilton, Winnipeg, Regina, Saskatoon, Calgary, Edmonton and Vancouver. All CMAs not on this list are located in Quebec, Ontario and British Columbia, with the exception of Moncton. Three more strata were formed by grouping the remaining CMAs (except Moncton) in each of Quebec, Ontario and British Columbia. Finally, the non-CMA areas of each of the ten provinces were also grouped to form ten more strata, for a total of 27 strata. Moncton was added to the non-CMA stratum for New Brunswick.

The survey frame was created using two different components: Lists of telephone numbers in use (both landline and cellular) available to Statistics Canada from various sources (Telephone companies, Census of population, etc.); and The Address Register (AR): List of all dwellings within the ten provinces. The Address Register (AR) was used to group together all telephone numbers associated with the same valid address. About 87% of telephone numbers available were linked to the AR. The records resulting from this linkage could possess more than one telephone number (grouped by the address). The other 13% of telephone numbers not linked to the AR were also included in the frame and each of them constitutes a single record. The combination of those two components results in the survey frame. The rationale for using all the telephone numbers (linked and not linked) was to ensure a good coverage of all households with telephone numbers. When more than one telephone number was attached to a record, they were sorted by source and by type of telephone number. The first telephone number was considered the best telephone number available to reach the household. Please note that for the remaining sections of this document, the word "record" will refer to the grouping of telephone numbers that consists of our sampling unit on the survey frame.In each frame, each record was assigned to a stratum within its province. A random sample without replacement of records was next selected in each stratum. All respondents in the ten provinces were rostered by telephone and interviewed by telephone or self-completed an electronic questionnaire. Households without telephones were therefore excluded from the survey population. For the 2015 GSS on Time Use, 87.4% of the telephone numbers dialled reached eligible households. An attempt was then made to conduct an interview with one randomly selected person from each household. To be eligible, a household had to include at least one person 15 years of age or older.

Question

1. According to what we saw in class and the readings, what type(s) of the sampling procedure is this (are these)?
2. Explain why it is this(these) type(s) of sampling procedure?
3. What are the benefits and limitations of using this(these) type(s) of sampling procedure over convenience sampling? (relies on individuals who are most accessible to the researcher, newscaster and now they select individuals who are closest in proximity)
4. What sampling error could result from this/these Statistics Canada sampling procedure(s)? Explain.