Which Polls To Trust?

I’ve heard from a few people really concerned about the polling in Florida, showing Gingrich down, and Romney up, which seems at least in part to be contradicted by the sense of things on the ground. I am going to show you how a poll can be twisted, and how you can’t trust them, and if you allow them to form your opinion, you can actually be manipulated.  Let’s start with a very popular poll, the PPP(Public Policy Polling) outfit that currenty shows Romney leading 40% to 32%.  

In this poll, when you read to the bottom of the page and examine the data, you discover the following:

“The PPP poll surveyed 387 likely Florida Republican primary voters on January 28th with a margin of error of plus or minus 5 percentage points.”(emphasis mine)

Ladies and gentlemen, when you see ANY poll with such a small sample size, it’s GARBAGE.  When you see any poll with a margin of error of more than 3%, it’s likely garbage, particularly if it has a small sample size.  For a state as large and diverse as Florida, any poll that doesn’t survey at least 1500 people, with a margin of error well below 4% is a poll intended to generate a headline and drive you.

Now, here’s their insurance: With a margin of error of 5%, if you add that to Gingrich, and subtract it from Romney, this poll may actually tell you nothing, because Gingrich could in fact be leading.  This kind of poll is GARBAGE as any serious analyst will tell you, and worse, when you see so-called analysts on television referencing such polls, you know they are LYING to you because as trained professionals, THEY ABSOLUTELY KNOW BETTER.

Sorry, I didn’t mean to “shout” at you, but I am disgusted with the media manipulation, as they try to create some false momentum in Romney’s direction. You should let others know this too.  387 people? Really? A margin of error of 5%??? That’s a potential 10-point swing in either direction, which gives them cover if their pushing doesn’t work out, but also helps them create a theme.

Don’t believe such polls. They are intended to mislead you. I have rejected four such polls from publication on this site in the last month because they were this egregious.

Advertisements