In this exercise you are presented with two short cases that involve marketing research. You are asked to identify and comment about the possible limitations of the research. The cases are accompanied by questions that will help to get your thinking started.

1. A marketing manager for a BMW dealership is trying to decide how many cars to order during the coming year to be sure to have enough on hand to meet demand. He decides that it would be useful to do a survey of customers to whom he has sold this brand in the last year. He wants to know how satisfied they are with their current car, and he wants to know how many want to buy another BMW from him in the coming year. He would also like to know if they could afford another expensive car so soon. He decides to have salesmen call the customers and ask the following questions:

(1) How do you feel about the car you bought from us? Are you very satisfied, or only moderately satisfied?

(2) Do you plan to buy another BMW from us during the coming year? Yes, you plan to buy; or no, you don't plan to buy.

(3) I have one final question, and your response will be strictly confidential and used only in statistical summaries with answers from other respondents. Would you please tell us your annual income? $_________________________________________________________

1. Do you think that customers will give a valid response to the second question? Why or why not?
2. Do you think that customers will give a valid response to the last question? Why or why not?
3. What is it about the way that the first question is worded that might keep the manager from getting valid information about how satisfied a customer really is? (Hint:Read the whole question several times carefully from the pointofview of different customers.)

2. A marketing manager for Bank of America wants to survey potential customers to see if they know about the bank's new investment services. An outside marketing research specialist tells the manager that for $5,000 the research firm can send out a mail survey to 500 people, tabulate the results, and present a report. He explains that the bank will need to provide a computer mailing list of people who have accounts at the bank-to save costs in developing the sample. He concludes by pointing out that the research will be quite inexpensive: "We will give you results from a representative sample of 500 people, at only $10 per respondent. And you can be confident with a sample of 500 that the statistics are accurate."

1. Is the proposed sample wellsuited to the manager's problem? Why or why not?
2. Is the researcher's concluding statement misleading? (Hint: Think about the response rate issue.) Why or why not?