Whether you’re doing market research or trying to figure out likely political winners, polls matter. An article in the journal Nature today has some interesting comments on the subject:
- By one estimate, only 35% of people reached by phone during the 2000 campaign answered pollsters’ questions, compared with 65% in 1985.
- The margin of error you hear about is an illusion. All it really guarantees is that the people sampled were statistically representative of the larger population. There are many other sources of possible distortions.
- One study found that it makes a difference whether the question “Who will you vote for?” comes at the beginning or end of an interview. The percentage of undecided voters dropped sharply when it came last.
No related posts.