The Monkey Cage

It’s worth noting that West/South/Midwest/West is a pretty broad breakdown, so the Republican party probably isn’t exclusively southern. For instance, I’d have to imagine Idaho/Utah/Wyoming remains the most Republican part of the country, but the bluer parts of the West (the coasts and the southwest) are far more populous, thus having the “West” overall look rather bad. Similarly for the Midwest.

But even so, even after you throw Mormon country and the Great Plains into the mix, the Republicans’ geographic base would seem to be rather narrow.

There is something deeply wacky going on in the fundamentals. For starters, how can 90% of Democrats disapprove of John Boehner, who I doubt 50% of Democrats have even heard of?

All I can figure is that there’s some sort of weird “You Lie!” effect going on this week.

The west includes Mormon land, Idaho, Utah and Wyoming which makes it look even worse for the Republicans.

Nevada also has an extremely high proportion of Mormons. But even apart from that, the “intermountain west,” between the Sierra Nevadas (with its counterparts in Oregon and Washington) and the Rockies is generally a socially conservative, capitalist libertarian area.

If, to take an extreme scenario, these regional discrepancies came to a civil war, I would expect the intermountain west to side with the South. This won’t show up the way the poll draws its regions because the coastal states far outweigh the states in the intermountain west in population.

Just to clarify, I’m Steve Benen, not Steve Bremen.

After finally understanding the difference between red (Republican) and blue (Democrat) states, now I’m confused with the colors used in the top graph. Perhaps the choice of colors reflects just how “blue” the Republicans are.

Keep in mind that the Kos/Research 2K poll includes non-voters in their weekly polling sample.

Note that Obama’s favorables/unfavorables in their Virginia poll was 45/48.

http://www.dailykos.com/statepoll/2009/9/16/VA/373

So how could his national numbers be at 55/38, when the Virginia numbers pretty much track the national numbers? The answer is that 17% of Kos’ national sample consists of non-voters while only 22% consists of Republicans.

PPP has found that polling showing such a small percentage to be wrong.

http://publicpolicypolling.blogspot.com/2009/04/republican-id.html

@Steve: My apologies! That’s what I get for blogging late at night… Your name has now been corrected in the original post.

@RepublicanComeback: Two points. First, if we’re tracking approval rating in order to get a sense of how popular the president is and/or how much public support he might enjoy for his agenda, why would we want to exclude non-voters? I’m not aware that it is common practice to exclude non-voters from measures of presidential approval - doesn’t this only come into play when trying to predict vote outcomes? Second, even if Daily Kos overestimated Republican unpopularity by X points in each region, that still shouldn’t affect the regional pattern of the distribution of support for the party. The only concern would be if Daily Kos was systematically underestimating Republican support differently in different regions of the country.

Joshua,

I have no issues with the Research 2K/Daily Kos state-by-state polling. They do not use non-voters for their state by state polling.

You bring up a decent point regarding non-voters. However, you did not respond to my argument as to how Obama could be at 45/48 favorables/unfavorables in Virginia and 55/38 nationwide. The answer has to be non-voters and while I acknowledge your point, my guess is that non-voters are irrelevant to politicians as they are what they are: people who don’t vote.

To me, it seems to be an attempt by Kos to pull up Obama’s numbers by including those who aren’t even the slighest politically engaged as evidenced by their self-identification as non-voters. If they were relevant, why not include them in the state-by-state numbers as well? The answer in my opinion is that Kos knows they are irrelevant because they don’t vote and would be insequential to polling them in a matchup poll.

With that said, wouldn’t you agree that Obama is struggling mightily in Virginia with a favorable rating of 45% in a state he won with 52% of the vote?

“You bring up a decent point regarding non-voters. However, you did not respond to my argument as to how Obama could be at 45/48 favorables/unfavorables in Virginia and 55/38 nationwide.”

There’s no “law” that VA numbers must track with national numbers.

Indeed, I see no reason to believe that they should, e.g., VA election haven’t consistently reflected national ones.

In fact, it is easy to understand why they wouldn’t: there’s no demographic, economic, or political reason why VA is a microcosm of the United States. VA is less black and less hispanic; it has a higher medium income; etc.

(see http://quickfacts.census.gov/qfd/states/51000.html)

Even if we thought that VA was a reasonable proxy for national numbers, we wouldn’t normally throw out he national numbers because hey didn’t track with the VA ones.

@RepublicanComeback: it strikes me as plainly pathetic that basically your saying “hey, the Republicans aren’t a regional southern party, look, they have Virgina!”

The fact that we should surprised Virgina disapproves of a Democrat president shows how low the Republicans really are.

I expect we may continue to see a hardening of regionalization, and maybe Virgina will fall on the red side of that.

Chris Lawrence, we Democrats all know who John Boehner is, but we pronounce his name differently than you do.

I would be cautious in using the DailyKos numbers as a yardstick. While I don’t know about the veracity of their overall numbers, I get the impression that their questions are asked in such a way as to make the Democrats look as good as possible and for the GOP to look as awful as possible.
The latest DK numbers show 41% favorable for the Dems and 23% favorable for the GOP. Other polling firms, however, don’t show the same massive spread. A recent Pew poll found the Dems at 49% favorable and the GOP at 40% favorable while CNN poll found the Dems 52% favorable and the GOP at 41% favorable.
Thus, I just don’t buy this argument that the South is the only region that likes the GOP. After all, Daily Kos is hardly an impartial outfit (although I know they claim their poll is)

In the first graph, it sure looks like three of the four graphs are giving your readers the bird.

Was that intentional, or a subconscious expression of your views about Republicans?

Steve: I can’t find the polls in question. Are you sure those are fav/unfav and not generic ballot?

One problem here is that the poll’s regional divisions don’t fit the actual political regions. So, two of the three Republican regions — the Mountain West and the Great Plains — are being lumped in with adjacent areas that are much more populous. Utah, Wyoming, North Dakota and Kansas are very Republican, but their signals are being swamped.

Doug M.

Also, the “regional party” notion doesn’t really fit the available data that well. For instance, there are 22 Republican governors; 7 are from the South (here defined as the former Confederate States) and the other 15 aren’t. This is not grossly disproportionate.

The “party of Southern White Guys” hypothesis doesn’t well explain the existence of Republican governors in places like Connecticut, Vermont, Minnesota and Hawaii. It also doesn’t explain why the state legislatures of (for instance) North Dakota, Arizona, Kansas, Wyoming and Utah are completely controlled by the GOP, nor why Republicans hold commanding leads dominate in odd places like the Michigan, Pennsylvania, and Ohio state senates. (N.B., when I say “dominate”, I mean “control thoroughly and won’t lose control any time soon; e.g., the GOP is 21-12 in Ohio’s upper house.)

You can say those are outliers and that the GOP is dominated by Southerners, but then at some point you have to step back and ask just what “dominated” means.

Doug M.