[kesg]

Rasmussen weights his samples using a 39% Dem 35% Rep 26% Ind breakdown. He believes that this will be the composition of the actual voter population on November 2 of this year. A large Harris party ID survey (6000 respondents) earlier this year found a simliar partisan breakdown among voters.

I am skeptical that this weighting is representative of party affiliation, especially among likely voters and especially if done earlier in the year (when Bush and Republicans generally were in a weaker position than they are today). More generally, I distrust polls that weight by party affiliation because the weighting destroys the randomness of the sample and thereby renders the poll unscientific.

[Josh in PA]

agree

[Illinois Rep]

"I am skeptical that this weighting is representative of party affiliation, especially among likely voters and especially if done earlier in the year (when Bush and Republicans generally were in a weaker position than they are today). More generally, I distrust polls that weight by party affiliation because the weighting destroys the randomness of the sample and thereby renders the poll unscientific."

That is very accurate. With Rasmussen's model there is NO way you'd ever have a landslide election being predicted by his polling. There is no way Bush could ever be polling significantly ahead of Kerry with his sampling demands. Yet landslides do happen (like this year possibly!!!)

[ScaniaBoy]

Hmm, isn't it Labor Day weekend now? (Or is that next weekend?)

For some reason Ras polls have alweays been worse for Bush on Sundays. If that has something to do with how easy or hard it is to get Dems vs Reps to answer questions during the weekend, may not that possible bias be increased during a public holiday?