A shoe company wants to conduct a survey of their customers to assess how much m

Question

A shoe company wants to conduct a survey of their customers to assess how much money they would spend on a new kind of shoe. They have a database with all 1500 customers who have bought shoes from them in the past year (including how much they spent in the past year), and they wish to sample 150 customers, but there are several opinions on how to conduct the survey. Employee A wants to contact the first 150 customers on the list who will respond to the survey. Employee B wants to randomly pick a customer among the first 10, and then to contact every tenth customer afterward. Employee C wants to take a 10% sample of customers from each letter in the alphabet according to their last name. Identify each of these sampling schemes and briefly evaluate strengths and/or weaknesses for each scheme. Which of these three schemes is best? Propose an alternative sampling scheme that incorporates the amount of money that each customer has spent on shoes in the past year.

Explanation / Answer

The sampling technique employed by A is termed as 'Cluster sampling', since here A just picked a cluster irrespective of any other parameter which might have an impact on the representativeness of the sample.

The sampling technique employed by B is termed as 'Systematic sampling', which is also a kind of probabililty sampling as long as the starting choice is completely random.

The sampling technique employed by C is termed as 'Stratified sampling', since here the employee first divided the whole population into strata using the alphabets in last name. Next from those stratas he randomly selected a 10% lot.

Of all these schemes, the one used by B is best since it is most close to being an unbiased sample because of the probability sampling used. Other sampling techniques would not produce truly representative samples.

Another sampling technique could be used in which the customers would first be organized according to the amount of money spent by each, and depending upon the magnitude of which each customer would then be assigned a probability of selection proportional to its size(money spent).

This technique is termed as Probability-Proportional-to-Size sampling.

Related Questions

Navigate

• Browse (All)
• Browse A
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.