Why Is Romney Leading Among Independents But Lagging In The Polls?
I noted on Monday that Mitt Romney seems to be maintaining a distinct advantage among self-identified Independents in both state and national polling. At the same time, and especially at the state-level, he seems unable to surpass President Obama in the top-line numbers. These facts have been frequently cited by conservatives as “proof” of their argument that the polls are oversampling Democrats and thus overstating the President’s performance. It appears, however that there may be another, more interesting, explanation for what’s going on.
Consider this chart from TPM’s poll tracker showing the trends in Party ID in those polls that report these numbers:
Look at what’s happening here. The number of self-identified Republicans has fallen while the number of self-identified “Independents” has increased to the point where it has either equaled or exceeded the number of self-identified Democrats since some point in 2011. Clearly, many of those “Independents” are Republicans who, for whatever reason, prefer identifying themselves as something other than Republican. At the same time, though, they are generally Republican in their political views and thus more likely to support Romney over Obama.
Every conservative commentator who decries party ID numbers in polls but then touts Romney winning independents in that poll should be forced to draw that chart 100 times on a blackboard.
Exactly.
The only other thing I would add is that the vast majority of Independents in the US are not being actively courted by either campaign — except in the swing states. So it’s entirely possible — at least in national polls — that some of these independents might have been persuaded to vote for Obama if they had been exposed to the same appeals as voters in swing states.
I’m not sure what the confusion on the conservative side is. Everyone knows a lot of Libertarian and (especially?) Tea Party voters are currently “Indie”.
Can’t imagine why people wouldn’t want to identify as Republicans. Aside from the fact that Republican is now synonymous with a range of modifiers from “nihilistic” to “mentally disturbed.”
If you go back to the ’08 election, the drop in Democrats is larger than the drop in Republicans.
I’ve often thought that it might be more useful if pollsters asked people to self identify by ideology instead … conservative, moderate, liberal. Then again, if that was the case, we’d probably end up with 75% of the country calling themselves “moderate”. 🙂
Weak tea, that. I don’t know who ‘TPM” is, but Gallup, Rasmussen and Pew all show significantly higher self-identification of Republicans. And they use those identifications in their poll weights.
Maybe you are having trouble accepting that genuine Independents prefer Romney?
ACORN strikes again!
@Jeremy Poole:
Correct. There were also more self identified democracts that republicans. Its been that way for most of the last 100 years (pew has a good interactive chart on this).
More recently however Republican affiliation has been on a decining path since the early 2000’s.
@DJ Drummond:
Gallup, Rasmussen and Pew all show significantly higher self-identification of Republicans. And they use those identifications in their poll weights.
You were pushing this lie back in 2008 to predict a McCain win. I see you’re still at the same stupid tricks.
@Todd:
So in other words you would like polls to do what polls already do?
Why are people so stupid when it comes to polls?
In today’s new Marquette poll of Wisconsin, the party ID breakdown is 29R/34D/33I and has Obama up by 8 points, leading to the usual rightwing idiocy about party ID. Yet the poll also clearly shows that by ideology, 42% described themselves as conservative/very conservative, while only 19% were liberal/very liberal. The remaining 34% considered themselves moderates.
Finally Doug comments on a phenomenon he is intimately acquainted with!
@michael reynolds:
What the hell did nihilists ever do to you that causes you to keep slandering them like this?
@mantis:
I wonder what excuse you will use a week from today?
You don’t need a Ph.D. in mathematics to figure this out.
Regarding that Party ID trend chart, it’s an amusing theory, in the sense that slapstick is amusing. There’s an obvious problem with that thesis.
If the phenomenon could be explained by disaffected R’s now self-reporting as I’s then ipso facto the I percentages in the polling surveys correspondingly would be higher. Capiche? It’s basic arithmetic. But that’s not what’s happening. The I percentages either are staying the same as historical turnout or they’re lower. The R and D percentages are fluctuating higher or lower with the spreads between D & R being the same or very close to the ’08 dichotomy (or in some extreme cases higher now for the D’s!). Only the I’s are not being reported higher, except for very rare variances.
IOW, the polling services are undersampling the I’s. The reasons should be obvious. And they’re engaging in wishful thinking about D vs. R turnout. Again, for obvious reasons.
Look, in the end none of this will matter. If Romney wins, even if by a larger margin than Bush in ’04, it was because of voter suppression, or Hurricane Sandy, or Diebold, or racism, or whatever. But if Obama wins, even if by a smaller margin than JFK in ’60, then it’s a “historic” affirmation of his policies and a “clear mandate” for him to impose the remaining parts of his agenda. Either way the chattering classes will be covered.
@michael reynolds: Actually, up in the corner of the Left Coast from which I come, conservatives have been identifying themselves as “independents” for about 30 or 40 years. The game when I was young was that conservatives would ridicule Democrats and party Republicans for voting straight tickets and being in lockstep with organized labor or country club rich guys like you. By identifying as “independent” conservatives could vote straight tickets “because the Republican candidate for this office just happens to be better.”
@DJ Drummond:
I wonder what excuse you will use a week from today?
You think my pointing out that you went on a nonsense crusade against Gallup and other major polling outfits in 2008, falsely claiming they weight by party ID and were doctoring their results to help Obama, is an excuse for something?
Guess what, dipshit? You were wrong then (and you deleted comments pointing out your falsehoods), and Obama won. What excuse did you use when the election showed how full of shit you were?
I guess it was all wasted effort on their part. Sure, they can think themselves independent, but in the stroke of a keyboard they can be assigned back where they belong.
@mantis: @mantis:
Hard to imagine how to answer that. Lots of lies, distortions and venom.
I guess I’ll just remind you that the results will come out soon enough.
@DJ Drummond: do us a favor and tell us again how the polls are all in the tank for the Dems because they are all located in blue states or something.
http://wizbangblog.com/content/2008/10/21/gallup-and-new-coke.php
Freakin’ comedy gold.
@Tsar Nicholas:
IOW, the polling services are undersampling the I’s. The reasons should be obvious. And they’re engaging in wishful thinking about D vs. R turnout. Again, for obvious reasons.
A sinister liberal conspiracy to destroy their own credibility? Makes sense to me!
@Barry Convex:
You’re dealing with a man who thinks he should use as a screen name theweak, incompetent, pussy-whipped, anti-semite who managed to lose an empire and get his family killed. So…
@DJ Drummond:
Actually, Pew’s most recent poll showed D 33, R 28, I 33. Thus Gallup and Rasmussen are on their own little island. And Gallup would probably be doing their usual thing of coming closer to the rest of the national polling through the week before the election if they were still conducting their survey right now.
Gallup has said that their LV screen only works in the 7-10 days before an election, and it has a history of giving really horrible swings and out of step with the rest of the world numbers before that. I suspect that Gallup’s final poll will move significantly in the direction of the national polling average and probably show something pretty close to a tied race nationally. At which point it will cease to be conservatives favorite poll and go back to being part of the evil, vast, left-wing polling conspiracy.
@DJ Drummond:
Pew DOES NOT show any significant uptick in Republican Party Affiliation for this year (unless it’s been in the last few months). Here’s their numbers on party ID back to 1939 laid out in an interactive graph:
http://www.people-press.org/2012/06/01/trend-in-party-identification-1939-2012/
@PogueMahone:
Thanks for the link. I chortled.
@PogueMahone:
Pogue, did you bother to read the article? Your derision sounds like you stopped at the headline,.
@DJ Drummond: This genuine Independent won’t have anything to do with Romney because of the company he keeps.
Gawd, that snotty little Paul Ryan jangles my last nerve.
The ongoing myth of the “independent” voter ….. It’s a mirage, there hardly any so-called independents. If there were as many “independents” as polling tells us there are we would have different congress than the dysfunctional right wing acid bath we have today … That is, unless these “independents” are as dysfunctional as the rest of the Republican base.
I don’t get it. In 2008 it looks like the D’s were at 36%, R’s were at 26%. R’s are at 25.2% and Ds are at 32.5%. Looks like the D’s took the bigger hit…