[Andrei Bulba]

Just asking a question here. I don’t believe the voter pool is only 23% republican. But are some of these pollsters claiming they randomize their method of contacting people, and then the party ID is whatever they happen to get from that “random” sample?

If so they can claim a clear conscience because, hey, that’s what people told us.

Or do they select the repub/ind/dem breakdown of their sample?

[erkyl]

Part of the reason is to demoralize Romney supporters and energize Obama supporters but they are also setting up the riots for when Romney wins by “obviously cheating” based on these polls.

It is heartbreaking to see what has happened to this country and the press in the name of liberal ‘progress.’

[InterceptPoint]

But are some of these pollsters claiming they randomize their method of contacting people, and then the party ID is whatever they happen to get from that “random” sample?

That is a good question.

The way it is SUPPOSED to work is that you poll X number of people, determine how many are R, D, I, Black, Brown, White, Young, Old, ... and then take this data and, with your own special sauce to get the right demographic mix you adjust the results based on your TURNOUT MODEL. In a nutshell the R/D/I split is the turnout model. This R/D/I split should be independent of the actual distribution of Republicans, Democrats and Independents who actually answer the phone for the poll in question. If it is done this way then there is no WEEKEND EFFECT, a false idea widely accepted here on FR that Dems poll better on weekends. OTOH, that belief could be correct if the pollster in question uses the distribution of R/D/I based on who answered the phone.

So what are they doing now? Who knows.

Well, as you have probably noticed, this so called turnout model is all over the map these days but is almost universally predicting a huge Dem turnout compared to the number of Republicans who will actually vote.

And what is the rationale for applying this Dem favored model? Answer: The hoped for effect of propaganda on the voting population.

That propaganda campaign worked even here on FR for a few days. I think it's fading as we dig out the facts about these polls.

[GeronL]

apparently at this point there are more registered GOP than Dems. So this poll is pretty whacked.

[LS]

I'm not sure how they are doing it. I'm guessing they first ask the last question, "What is your party identification?" When they get a % of Ds they want, they stop taking any more responses from Ds. So it HAS to be predetermined what their mix is.

They would not just call 1000 people and take whoever answered the phone, because you might get 1000 Ds or 1000Rs by accident.