Reidblog [The Reid Report blog]

... For the 2006 elections, Democrats cannot increase their advantage among independents any further than they already have. This is the largest Democratic lead among independents in 24 years, and historically is only clearly surpassed by the advantage they held among independents in 1974.

Given this lead among Independents, there has to come a time when Democrats realize that success in this election depends less on continuing to target and appeal to Independents, and more on building a political machine that can make their current appeal and potential majority into a reality at the ballot box. At the same time, there needs to come a point within the progressive activist base when we realize that in our lifetimes it is entirely possible that there will never be a better opportunity than 2006 to wreck permanent damage on the conservative movement and all for which it stands. Pass up this chance, and the next time an opportunity of this level comes around there is a good chance you will be either dead or retired.

...I suppose I should be a smart, non-vindictive blogger and trumpet this as good news. After all, it does not really matter how different polling firms compare with each other. Every polling firm has a different "house effect" that skew in one direction or the other on average, and so the more salient results are found within the historic trends of any individual polling firm. Thus, it isn't really important how Gallup's numbers compare to Harris, Pew, or the National Election Survey, but rather how Gallup's numbers compare to themselves. In this regard, for Gallup to show a shift in favor of Demcorats is undeniably a good thing for Democrats.

However, I am a vindictive blogger that holds long-term grudges against a small number of people and organizations, and as such I would like to point out how the only thing historic about this shift is probably that Gallup is now at least somewhat in line with the other three major polling organizations that conduct major studies of national partisan self-identification. While Gallup showed a very narrow one or two point Democratic lead for 2005, Harris, which polls about 6,000 people a year, showed a 6-point Democratic margin for 2005. In late 2004, when Gallup was showing a 2-point edge for Republicans, the National Annenberg Election Survey of over 67,000 registered voters showed a 2.8% edge for Democrats (PDF). In 2004, when Gallup was showing a 2-point edge for Republicans, Pew, which polled 19,000+ registered voters, showed a 4-point edge for Democrats (I can't find Pew info on 2005).

In other words, no matter how many people they poll (roughly 8,000 every three months), Gallup has consistently measured the country about 5% more in favor of Republicans than the other three major pollsters who conduct huge, national studies of partisan self-identification. Rather than trumpeting a historical shift that was only historic because their data from 2004 and 2005 disagreed with everyone else's, maybe Gallup should develop some sort of explanation as to why their random sampling methodology consistently turns up more Republicans than every other major public, political polling firm in the country.