Most people have heard of the unemployment rate, but not so many know where it comes from. Does the rate simply represent the number of people claiming Employment Insurance (EI)? It turns out that that would be an underestimation of the number of people unemployed, since many people are unemployed but ineligible for EI. Instead, Statistics Canada conducts the Labour Force Survey, interviewing people to find out their employment status and then estimating the unemployment rate for the whole country. During the second half of every month Statistics Canada analysts survey about 50,000 households, analyze the responses, and report the results. The most widely publicized number from this survey is the unemployment rate, but the survey covers much other information; for example, shifts of employees from one industry to another, hours worked, and demographic information about employees including age, sex, marital status, education level, and province or territory of residence. How would you design the Canadian Labour Force Survey?

• What is the population of interest?

• Why might it be difficult to select a simple random sample from this sampling frame?

• What sampling technique would you use to ensure that we have a representative sample of people from each province and territory and from the demographic groups described above?