Question: Are Rasmussen and Gallup polls automated or done by a human?

I'd like to know because I am trying to factor in the Bradley Effect. If they are automated then we can almost eliminate the Bradley Effect because secrecy is greater and thus those polled are more apt to reveal their true opinion.

I am one of the folks they poll and yes, I confess- I am human!

Was it an automated, "touch tone" process? Or did an actual person ask you the questions?

but my understanding is they are live people by phone.

Polling over the phone (with a live human) actually does involve a fair amount of personal detachment that may be sufficiently "anonymous" for a potential voter to reveal what they really think.

Historically, the Bradley Effect was determined by exit polling where voters talked to *actual humans* and expressed who they voted for. In that case the actual face to face encounter probably created more anxiety and increased the chances of telling a white lie to hide "guilt".

Polling over the phone is pretty accurate in that people tend to be more honest. The few areas where it is tough is when you use aided recall.

Ask a Chicago Cubs fan if they watched the game last night. They'll say "yes" even if they were in a different country last night. A "yes" supports the team. Also, research has shown, that people tend to say "yes" to yes or no questions because we typically do not like to disappoint others. A yes can be viewed usually as supportive no matter what the question. Last, Yes answers can usually move an interview along more quickly and some people will do what they can to get to the end of a survey.

Overall the magic is in the way the questions are written and the talent of the interviewer.

Note, these 2 polling calls were for 2 separate elections.

automated, but seemed to go on to different questions depending on my answer - took about 5-10 minutes.

and it was a human. She asked me about 200 questions.

People can latch onto any one of dozens of real/manufactured reasons for not voting Obama. There is no public expectation that anyone vote for Obama; on the other hand, there exists the very real pressure on people to conceal any racist leanings.

In actuality, it is more likely that McCain/Palin feel the negative Bradley Effect. Some white people may feel self-conscious about expressing public support for a black man. What they'll tell a human pollster may be at odds with what happens when they're in a private voting booth.

Either way, I doubt the Bradley Effect is good for more than a 1 or 2 point variation from the polls.

Jes' guessin'.