2. Gallup conducts its polls by telephone, so people without phones are always excluded from the Gallup sample. In order to estimate the proportion of all U.S. adults who plan to vote in the upcoming election, Gallup calls a random sample of 500 U.S. adults and constructs a 95% confidence interval based upon this sample. Does the margin of error account for the bias introduced by excluding people without phones?
(A) Yes, the error due to undercoverage bias is included in Gallup's announced margin of error.
(B) Yes, the margin of error includes error from all sources of bias.
(C) No, the margin of error only accounts for sampling variability.
(D) No, but this error can be ignored, because people without phones are not part of the population of interest.
3. Which of the following is the best way for Gallup to correct for the source of bias described in the previous problem?
(A) Use a better sampling method.
(B) Select a larger sample.
(C) Use a lower confidence level, such as 90%.
(D) Use a higher confidence level, such as 99%.
