Refer to Exercises 1 and 2. Suppose that a local newspaper conducted a survey during the next week, based on random-digit dialing of 1400 residents of the area, and found that 20% of them reported having seen a UFO on the evening in question.
a. What is the margin of error for the survey?
b. Give an interval of values that probably covers the true percentage of the population that saw the UFO.
Exercises 1
Refer to Exercise 2. Acknowledging that some bias is inevitable when information is required on such short notice, which of the four methods would you recommend that the television station use to get the information it desires? Explain.
Exercise 2
One evening around 6 P.M., authorities in a major metropolitan area received hundreds of phone calls from people reporting that they had just seen an unidentified flying object (UFO). A local television station would like to report the story on their 10 P.M. news and wants to include an estimate of the proportion of area residents who witnessed the UFO. Explain how each of the following methods of obtaining this information would be biased, if at all:
a. A "quickie poll" taken between 7 P.M. and 9 P.M. that evening, using random-digit dialing for residents of the area.
b. During commercial breaks between 7 P.M. and 9 P.M. that evening, providing two numbers for viewers to phone: one if they had seen the UFO and one if they had not.
c. A door-to-door survey taken between 7 P.M. and 9 P.M., based on a multistage sampling plan, in which 10 city blocks (or equivalent area) were randomly selected and interviewers knocked on the doors of all homes in those areas and surveyed those who answered the door.
d. An e-mail survey sent at 7 P.M. to all e-mail addresses for local residents served by a local Internet service provider in the area, using responses received by 9 P.M.