Comment

Overnight Open Thread

1252
iceweasel7/31/2009 9:14:50 am PDT

re: #1197 VioletTiger

In what way are they biased and flawed?

A few ways. Again, you have to evaluate on a poll-by-poll basis, but here are the flaws to look for, and ones that Rasmussen repeatedly has:

Sample selection: they poll ‘likely voters’, not all voters, and they define ‘likely’ in such a way that it excludes people who missed the last presidential election vote, or for whom that was their first vote. This automatically excludes a lot of people, and the people excluded tilts the sample selection to the GOP (because Obama had such an overwhelming edge with young voters and first time voters). This is also why all Rasmussen polls will show Obama approval trending lower than other polls.

Methodology: The questions asked and the framing of the questions. Time after time in a Rasmussen poll the questions are slanted in such a way as to produce results that will accord with GOP positions. This isn’t going to apply to every Rasmussen poll, the way the first objection does, which is why each Rasmussen poll needs to be evaluated on an individual basis. Unfortunately it applies to far too many of them.

I’ve just learned over time that they’re untrustworthy because these errors keep popping up.