I
don't necessarily question the results of the poll. I'm 95% certain that those are the actual results that they obtained + or - 3%. Extrapolating those results to the rest of the population though, is rife with problems. Gallup is asking us to believe that their sample of 1000 people
was random and
does accurately represent the feelings of the entire population. I find that hard to believe based on my experience.
Drawing conclusions from any data set is always difficult and one must be very careful not to "read too much" into the data.
I'm not around gun-people most of the time. (I work in a public school that is filled with liberals) Nobody I know (except for one) is talking about gun control. Everybody here is talking about how the carnage could have been limited by decisive action like on flight 93. At least one other teacher has her CWP and others seem open to the idea of trained individuals being allowed to carry on campus.
Gallup makes no pretence of conducting scientific research. They are a political polling organization and work for whomever pays them. I've done too much real scientific research to believe that a survey accurately represents any population other than the sample that was polled. There
are ways to correct or control for bias but any statistical transformation of the data has an associated set of problems. Phone surveys are especially notorious for skewing results for the reasons I mentioned in my first post. The site also says that these questions were part of other polls. What were the questions like on the rest of those polls and were they likely to introduce bias?
Having said all that, I'm not resting on my laurels. I'm voting in on-line polls, writing my letters, making my phone calls and sending my money to the NRA.