Thursday, May 20, 2010

Statistics?? Survey?

Suppose a recent survey shows that 75% of a sample of Florida drivers would like the speed limit reduced on highways. The margin of error is stated to be 3%. While the statistics may seem conclusive, what facts would you want to know about the survey before buying into the results?

Statistics?? Survey?
a) who answered the survey - random selection or not random selection


b) are we talking about new drivers or experienced drivers


c) was survey administer across florida or only in major cities of florida?
Reply:You may be interested in how the people in the survey was selected. Was this a random sample or not? Was there a gender or age group or race that was more likely to be chosen than others due to the type of sampling used?





Were the people in the sample asked the question in a way such that it did not induce any type of response? For example, if the interviewer asked





"In order to reduce the possible of traffic fatalities, should the speed limit be reduced?"





then the question may lead to more yeses.


No comments:

Post a Comment