Akkam’s Razor

This is a nice single post that highlights a few of the problems with the medias love affair with polling, their lack of understanding of some key elements, and some of the inherent problems keeping some of the polls from being scientifically sound.

As a a Newsday article points out, some polls are done using criteria that excludes LARGE portions of the population. For example, cell phone users (those where cells are their primary phones) are not counted. Secondly, look at the criteria for a likely voter. It’s not hard to imagine large amounts of individuals who would be excluded from registered voter polls, and would also explain the discrepancies between Registered- and Likely-Voter polls.

Popularity: 1% [?]