^^You're probably the reason a small percentage of "atheists" believe in God according to surveys, moorish =P.
Still, if you have a more reliable poll, I'm all ears. Also, 1000 people isn't a bad sample size. It's about what you want from a poll for the results to stick. As for the fact that it's a phone survey, in the year 2013, where most people have a phone, I don't see it as much of a problem. No poll is perfect. A mail survey likely would yield similar results, an online survey would be less reliable due to it only measuring people with high levels of interest and those who have PCs. Quite frankly, this poll could be a lot worse.
Also, if you look deeply into the polling data, they mention that the views of those who said God only no evolution believed the world was made in the last 10,000 years, that was part of the question.
Also, the fact that it doesn't change much means that the results are rather consistent. Meaning either their polling methods are reliably wrong, or their conclusion is reliably correct and there's not much of a trend taking place.
Also, what other stats? A quick google search brings up a ton of citations of this poll. I think this the main poll cited.
EDIT: Pew poll still cites about 30% are creationist.
Although honestly, I actually trust their methodology less. They mention doing some things for certain polls like oversampling or using online surveys. Overall, their methodology isn't much different than gallups' though, their main method is phone surveys...that's probably the best way to do things.
Yeah I missed how it mentioned the 10,000 years thing, but I swear I've seen statistics that say the majority of Christians don't deny evolution, which conflicts with this data that says more of them do.
You are the universe Expressing itself as a human, for a little while
That's still not a very good way to do a poll though, since you're only polling people willing to do a phone survey.
It's not exactly a sample size that you can control. Just "adults who say yes to a phone survey" and you do it enough times to get enough results. And if I'm reading it right, it's only 1,000 people.
In addition, another problem I've read about recently, and particularly in regards to Gallup polls, is that they only dial land lines. Used to be that was ok, but in the age of smart phones and skype and whatnot, it's starting to select for a certain demographic. Younger and more tech savvy people are decreasingly likely to have a land line, and thus are increasingly likely to be excluded from these polls.
I don't know if the skew is significant yet, but it is something to be aware of with phone polls.
Time flies like the wind, and fruit flies like a banana.