Question: DATA to DECISION

Critical Thinking: Questioning survey results

Because surveys are now so pervasive throughout our society, each of us should develop the ability to think critically about them. We should question the process of selecting survey subjects, the wording of the question, the significance of the results, the objectivity of the survey sponsor and the group conducting the survey, as well as other issues. In a Gallup poll of 1038 randomly selected adults 18 or older, each subject was asked this question: "In general, how harmful do you feel secondhand smoke is to adults?" The subjects were given these choices: very harmful, somewhat harmful, not too harmful, not at all harmful. Here are the results:

Very harmful: 52%

Somewhat harmful: 33%

Not too harmful: 9%

Not at all harmful: 5%

No opinion: 1%

Analyzing the Results

1. The above results are recent, but a 1994 poll resulted in a 36% response rate for the choice of "very harmful." Use a 0.05 significance level to test the claim that the more recent percentage of "very harmful" responses is greater than the 36% rate from 1994.
2. Can a hypothesis test be used to verify that the 36% rate from 1994 has not changed? Why or why not?
3. If, instead of a random selection of subjects, Prevention magazine obtains results by publishing the survey that readers could return by mail, how are the hypothesis test results affected?
4. Does the wording of the question appear to be neutral and without bias? Provide an example of wording that would be somehow biased.

Price: $2.99

Solution: The downloadable solution consists of 3 pages
Deliverable: Word Document