I thought that was how all polls are conducted by a random selection unless you are used to your trusted News Outlets to just pick and choose who ever they think will give the a better outcome on whichever side they are biased on. Are you saying that FOX NEWS is biased along with the money other news organizations that a reporting the same statistics as FOX and has something against Romney so they have an agenda when it comes to polling?
They most likely ( just my guess) oversampled on Hispanics because they want to see their specific view on the campaign since the Hispanic vote is a big deal for both Romney and Obama.
Originally Posted by SkylarCruzWantsYou
No, polling performed in order to predict the outcome of an election, is absolutely not supposed to be done based on a randomly selected sample group.
Legitimate pollsters are trying to only poll the people most likely to vote; that isn't acheived by random selection. It doesn't do the pollster any good to know how people intend to vote, if in fact, they don't intend to vote.
In addition, legitimate pollsters make an effort to poll a sample group that is demographically representative of people who actually turn out to vote. They try not to interview too many old people, too many young people, too many white people, too many Democrats or too many Republicans, etc.
It's very easy for a pollster to get any result he wants, simply by oversampling a certain group. You can't just poll an NRA meeting or a NAACP meeting, and assume that's going provide an accurate representation of the national vote.
Poll results are meaningless, unless the sample group is fairly chosen. Many of the recent polls are obviously rigged.