No matter how the sample ratio has been modified (turnout %, etc.), it is difficult to imagine a national sample with a 7% greater representation by Republicans. It is likewise difficult to believe that the Democrat and Independant samples are equivalent. IMO the pollsters are busy trying to favor the Republicans by 10%+ in the sample so that they can run a "corrected" poll in a few weeks showing a "precipitous decline" in GW's numbers following recent "events and disclosures" forthcoming by the Dimo establishment.
Dear Mad_Tom_Rackham,
As I pointed out in #8, it doesn't appear that Newsweek used a weighting of 38% R, 31% D, 31% I. That's just the raw numbers.
sitetest