Gastwirth (1988, p. 507) describes a court case in which Bristol-Myers was ordered by the Federal Trade Commission to stop advertising that “ twice as many dentists use Ipana [ toothpaste] as any other dentifrice” and that more dentists recommended it than any other dentifrice. Bristol- Myers had based its claim on a survey of 10,000 randomly selected dentists from a list of 66,000 subscribers to two dental magazines. They received 1983 responses; with 621 saying they used Ipana and only 258 reporting that they used the second most popular brand. As for the recommendations, 461 respondents recommended Ipana, compared with 195 for the second most popular choice.
a. Specify the sampling frame for this survey, and explain whether you think “using the wrong sampling frame” was a difficulty here, based on what Bristol-Myers was trying to conclude.
b. Of the remaining four “difficulties and disasters in sampling” listed in Section 4.6 (other than “using the wrong sampling frame”), which do you think was the most serious in this case? Explain.
c. What could Bristol-Myers have done to improve the validity of the results after it had mailed the 10,000 surveys and received 1983 back? Assume the company kept track of who had responded and who had not.