1. Gallup Polls When Gallup and other polling organizations conduct polls, they typically contact subjects by telephone. In recent years, many subjects refuse to cooperate with the poll. Are the poll results likely to be valid if they are based on only those subjects who agree to respond? What should polling organizations do when they encounter a subject who refuses to respond?

2. IBM Survey The computer giant IBM has 329,373 employees and 637,133 stockholders. A vice president plans to conduct a survey to study the numbers of shares held by individual stockholders.

a. Are the numbers of shares held by stockholders discrete or continuous?

b. Identify the level of measurement (nominal, ordinal, interval, ratio) for the numbers of shares held by stockholders. c. If the survey is conducted by telephoning 20 randomly selected stockholders in each of the 50 United States, what type of sampling (random, systematic, convenience, stratified, cluster) is being used?

d. If a sample of 1000 stockholders is obtained, and the average (mean) number of shares is calculated for this sample, is the result a statistic or a parameter?

e. What is wrong with gauging stockholder views about employee benefits by mailing a questionnaire that IBM stockholders could complete and mail back?