Four sampling strategies have been proposed to help a bank determine whether customers favour opening on Saturdays versus keeping branches open longer during the week. For each, discuss what kind of sampling strategy is involved and what (if any) biases might result.
(a) sponsor a commercial during a TV program, asking people to dial one of two numbers to indicate which option they prefer.
(b) hold a meeting at each branch and tally the opinions expressed by those who attend the meeting.
(c) randomly select one day at each branch and contact every customer who visits the branch that day.
(d) go through the banks customer records, selecting every 100th customer. Hire a survey research company to interview the people chosen.
a) Voluntary response sampling. This represents one of the non-probability methods, where instead of randomly selecting individuals there is a non-random aspect. It would bias results towards individuals who watch television, and those who call in phone numbers.
b) Voluntary response sampling. This is comparable to the previous method, but occurring in-person instead of through the media. Again, individuals who like to participate in events would vary systematically from the average, thus biasing results.
c) Cluster sampling. Each branch serves as a region, then an individual is randomly taken from it. This should not have that much bias, given that it is a probability sampling method. It would be biased against individuals who do all of their banking online or through the telephone, however.
d) Systematic sampling. This is also a probability sampling method that is not too subject to bias. Given that the sampling frame comprises the entirety of the bank's customer database, it should have fairly unbiased results.