The trouble with polls

If you watch talk shows or listen to talk radio you hear some strange and confusing things.

The Democrats have a poll that shows that 67% of the people want a public option in the Health Care plan. The Republicans and the Tea Partiers have a poll that shows that 52 % of the people don't want any Health Care plan at all. How is it possible that both of them are right?

Simple. It depends on who is polled and how they are polled.

Take Jack Cafferty's poll. He editorializes before he asks the question which automatically skews the answers he gets. Take Ed Schultz's poll. His audience will come up with answers that he expects. The same goes for most of CNN's polls and the O'Reilly, Glen Beck, Rush Limbaughs polls. The same goes for Miringoff, Quinipiac, Roper, Gallop or anyone else.

They all have a list of whom to poll. After a few polls they have a pretty good idea of how a person will answer. They know who is Conservative, who is Liberal, and who is independent. So if they are being paid by someone who wants to show a  Conservative or Liberal leaning, all they have to do is call those people who will give them the answers they are seeking.

They will argue that having that list shows trends, which is somewhat true. If 10% of the people they call show that they are changing their position it is what some organizations want to know.

Unfortunately it also allows them to skew the polls.

You pay me enough to poll and I will get a poll that shows some people believe that Karl Rove and Rush Limbaugh are the same guy. I will also get you a poll that says that most people do not believe any polls, because they have never been polled.

If you want to judge a poll, find out who paid for it. Think of what they are looking for, and how they benefit from the poll.

The problem with some polls is that we elect our officials sometimes because of poll numbers and they vote on legislation based on polling. Polls can be beneficial, but remember to question why they're done.

No poll can show what the American electorate really thinks. They are just samples and should be treated as samples.

That is why they are sometimes dead wrong.