Everytime The Drudge Report hysterically trumpets a poll showing that McGrumpy might win this thing, just head over to FiveThirtyEight where Nate Silver will calmly explain why the poll in question is an outlier or has a biased sample. He will do so using fact, logic, and science. Like this.

Trick or Treat

Do you spook easily?

Judging by the response in my inbox, some of you do.

Matt Drudge is touting the results of a one-day sample in a Zogby poll, which apparently showed John McCain ahead by 1 point.

There are a couple of significant problems with this.

The problems are numerous. It’s a one-day poll, and as these can be  unreliable. Also, it’s Zogby poll and uses “nonsensical party ID weightings which mean that his weighting process involves making numbers doing naughty things that they usually don’t like to do.”

Plus, six out of seven other polls taken at the same time had Obama up, thus making the Zogby poll an outlier. Which would not be surprising if you are doing naughty things with the numbers.

In short, Zogby polls are not reliable. I’m guessing Drudge knows that too.

No Comments »