PPP/DailyKos weekly poll: Obama+5 nationally (likely voters)

[ChrisFromNJ]

You asked "if you can call it that." The answer is: No, you can not. Trying to cover one's ass with "It is irresponsible not to speculate" does not make the speculation any less foolish.

You have repeatedly demonstrated that you do not understand polling, particularly party ID.

1) It is not a demographic characteristic
2) It is a dependent variable of voter intent
3) The relationship is weak

There aren't "more" or "less" Democratic of Republican electorates, since party identification is an attitude, not a characteristic. Party ID breakdowns reflect how the sample intends to vote - it does not drive it. And even those two variables don't correlate very well.

Bonus question: Which election had a 38D-35R party split?

A. 1984
B. 1988
C. 1992

Listen, there's no need to act like the jedi of poll interpretation. The answer to your question is 1988 according to the Washington post http://www.washingtonpost.com/wp-srv/politics/interactives/independents/data-party-identification.html, but comparing 1980s party identification (or even party registration) to today is pointless. We don't have a consistent measure of party ID/registration across the country, but the data we do have tells us there were a lot more people identifying as Democrats in the 1960s and 1970s than today. Even  in 1984, when Regan won by nearly 20 points, party ID in the exit polling was D+2.

Looking at the last several elections, you can certainly see a correlation between party ID in the exits and the results. In 2002, when Republicans had a great year, party ID was near even, or R+1. In 2004, when Republicans won their first MAJORITY of the popular vote since 1988, party ID was even. In 2006, when Dems had a great year, turnout was D+3. In 2008, when Dems had an even better year, it was D+7. In 2010, when Reps had a great year, party ID was back at parity again. So you can see, as party ID approaches parity, the results bode well for the GOP. As they approach D+7 (2008) and above, the results imply landslide status for the Democratic party. I don't see what is so wrong with discussing this.

Your contention that there are not "more" or "less" Democratic or Republican electorates is an opinion, not a fact. You can read an exit poll and see that 39% of voters identifed as Democrat in 2008, and 32% identified as Republican. Political and poll analysts frequently cite the MORE Democratic electorate of 2008, often pointing specifically to the party ID in that years exit polling to make their point.

I'll repeat that there is nothing in my post that demonstrates I know nothing about polling, ole wise professor. The party ID of the poll is D+5 (a fact). Obama leads by 5 (a fact). The party ID of the 2008 electorate based on exit polling was D+7. A fact. Nothing about pointing out those facts implies you know nothing about polls. Stop being the "poll analysis" police.

And since you're so wise, please do tell us what we are allowed to discuss when looking at poll internals? You know, just so we can know what we're allowed to talk about around here.

I think what you just showed us is that election results drive Party ID, not the other way around. As of right now, we do not know what the results of the 2012 election will be. However, we have numerous pollsters making projections that are very fluid, and not sticking to a static party ID that projects conventional thinking onto the electorate. I think if you say that "there is no way the electorate is going to be just as Democratic as it was in 2008", you are projecting your own feelings into the race. Based on live interviews in the field, pollsters may be seeing something different.