Associate
- Joined
- 6 Mar 2011
- Posts
- 140
- Location
- Scotland, UK
Actually, you can do that. A great deal of statistics deals with such things, where you know nothing and you want to know how many samples (ie people to ask) to take in order to have X % confidence your results are within Y% of the true value.yes extrapolating the answers of 1000 people and applying that to over 300 million.
yeah that's going to be reliable.
1000 people gets you something like 90% confidence you're within about 4% of the true values.
However, and this is what sticks a fork in so many surveys, the sample must be random. As pointed out, 1000 p[eople from the Bible Belt isn't a random sampling of the population as a whole. What the survey tells you is that ~1/3 of people in that area think that.
1/3 is a high number but a higher fraction don't even accept evolution. I watch The Atheist Experience on Youtube a lot and it has helped restore some of my hope for the US. I very much recommend it to anyone wanting to see the standard religious nonsense be slapped down with intelligence, reason and occasionally "What is wrong with you?!". Its a shame they are an island of sanity in a sea of ignorance.