wow 'Seriously' you are a bit thick aren't you?
No, actually pollsters call numbers randomly and amass a pool of adult responses. They then reweight the poll to census data. So if they get 60% women responding they adjust the weight. Bear in mind that over 90% of respondents do not agree to take the survey. This issue of response rate has become a big problem. There is some evidence that there is some response bias after certain events like conventions where those of a particular political persuasion are more enthused and more likely to agree to be polled. But again the pollster does not map anything out before they begin calling except to make sure their calls are balanced geographically.
The fundamental misunderstanding with the new attack on polling is a belief that pollsters predetermine what the party ID makeup will be. The only pollsters who do that are Rasmussen and the Battleground poll.
Turnout does not equal census data as all races, ages and even gender do not vote equally. Women vote more than men. Whites vote more than Hispanics. Old people vote more than young people.
At the end of the day, no matter how many phone calls you make, the data is never perfect.
There has to be some sort of judgment call made and the data has to be raked to mirror some reasonable expectation of what the electorate will be.
Do you have any evidence that pollsters are getting race and age demographics wrong?
