INTL 725 Research Methods for Trade & Development

Question 1: Probability sampling designs You are a researcher at Nova Bank that has 100 branches across Canada, with a total of 100,000 clients and 2,000 employees. The bank president shows you the CIBC Poll: Am I saving enough to retire? (http://cibc.mediaroom.com/2018-02-08-Am-I-saving-enough-to-retire-Vast-majority-of-Canadians-just-dont-know-CIBC-poll) in which “an online survey was conducted among 1,523 randomly selected Canadian adults who are Angus Reid Forum panellists. The margin of error… is +/- 2.5%, 19 times out of 20.” The bank president asks you to explain 4 ways of selecting a random sample of 2,000 from the list of 100,000 clients, for a similar Nova Bank poll on retirement savings. a) Explain each of the 4 probability sampling designs for this Nova Bank poll to select a sample of 2,000 from the list of 100,000 clients, and the advantages and disadvantages of each design. • Hint: for stratified random sampling, use gender and age groups (see CIBC Poll); for cluster sampling, use geographical groups (cities) b) Use SurveyMonkey (https://www.surveymonkey.com/mp/margin-of-error-calculator/) to calculate margin of error of this survey with 95% confident (which equals 19 times in 20). Explain what it means for Nova Bank poll by “The margin of error is +/- ___%, 19 times out of 20”.  Hint: Read this article (https://www.isixsigma.com/tools-templates/sampling-data/margin-error-and-confidence-levels-made-simple/). Question 2: Non-probability sampling designs The bank president also asks you to explain 3 ways of selecting a non-probability sample of 200 from the list of 2,000 employees for a qualitative survey on the challenges and opportunities for managers who are women and/or members of visible minorities to advance in the banking industry. Explain each of the 3 non-probability sampling designs for this Nova Bank qualitative survey, and the advantages and disadvantages of each design