Presidential Pollsters who attempted to tell us the results of the presidential election proved to be as ignorant as a box of bricks. Even the day after the horrible wrong polling the pollsters still don’t get what they did that was wrong. They keep explaining the issues that made their polls so wrong but it was the polling, not the issues.
“Most of the press and folks in DC were science deniers when it came to this election,” said veteran GOP operative Curt Anderson, an adviser to a pro-Trump super PAC. “Even in the face of polls that showed it very close, they all said that Trump had almost no chance. It was because they couldn’t imagine it happening.”
Some pollsters blamed the people for not telling them they were going to vote. Others explained that the Comey announcements threw off their polls. Others said they should have polled earlier or later. The problem is with them, not those they polled.
A simple drive along the highways would have shown them the massive imbalance in Trump political yard signs compared to the small number of Hillary signs.
Some blamed white racism even though the election was between two white people. Never mind that they say, Hillary was the Obama candidate and Obama is a black man so the whites who voted for Trump were racist. This racism nonsense is getting so tiring and so trite. It’s not racist to vote for someone who is the same race or not to vote for them if their race is different. Nature isn’t racist and nature decides the race of each person. Different skin colors are natural, not racist. Or maybe without evidence we should accept the theory that white people are racist and be done with it. So what?
But polls are supposed to make people smarter about the unknown. In the case of the 2016 presidential election they made us dumber. They misled us. They mis-informed us. The conclusion must be that presidential polling is inherently wrong.
But it doesn’t have to be and it shouldn’t have been so misleading. Polling is part of the science of Statistics. The problem is: polling that fails to correctly predict the results of the voters days or weeks ahead of the election should never be trusted. Remember that the next time someone tries to predict the result of a presidential election.
A huge problem for pollsters is human variability. A person may believe “X” to day and “NOT X” tomorrow. People change their minds based on a lot of variables so using data over a period of time will frequently be corrupted because of humans being human. It shouldn’t surprise and there are procedures to eliminate the incorrect data but it’s another factor that should be accounted fo in presidential polling.
The presence of an interviewer can affect the answers. Likewise the poll itself may be viewed by the subject as a test of his or her ability to provide the correct answer instead of simply sampling their unbiased answer.