17
   

Get yer polls, bets, numbers & pretty graphs! Elections 2008

 
 
nimh
 
  1  
Reply Fri 25 Apr, 2008 08:55 am
Great maps... very clearly lifting out a geographic pattern that could easily be lost in state-by-state approaches:



Quote:
Pennsylvania Confirms Hillary's Appalachian Prowess

The Electoral Map.com
April 24, 2008

Hillary Clinton's sweep of Pennsylvania hill country was the latest indication that she is much more popular in Appalachia than her rival Barack Obama. Hillary has been has been posting big wins in congressional districts like Virginia's Ninth and Ohio's Sixth and racking up huge margins in Appalachian counties from the Tennessee River to the Susquehanna.

I've noted before that Clinton is dominating hill country, and J-Mart has written about her Scots-Irish advantage and Michael Barone has discussed her appeal to Jacksonians.

To test these theories, I identified the congressional districts I consider to be in Appalachia, and then mapped out the electoral results. From south to north, I included Alabama's 5th; Georgia's 9th; South Carolina's 4th; Tennessee's 1st, 2nd, 3rd and 4th; North Carolina's 5th, 10th and 11th; Kentucky's 5th; Virginia's 6th and 9th; West Virginia's 1st, 2nd and 3rd; Maryland's 6th; Ohio's 6th; and Pennsylvania's 4th, 5th, 9th, 12th, 14th and 18th.

North Carolina, West Virginia and Kentucky have yet to vote, but if the electoral maps of the rest of Appalachia are any indication, Hillary can expect huge wins in the Appalachian pockets of these states.

Electoral Map of Appalachia In this map, Clinton is red and Obama is blue. The darker the shade, the higher the margin of victory.

http://theelectoralmap.files.wordpress.com/2008/04/04-24-appalachia.jpg

Electoral Map of Appalachia (Race42008.comhttp://theelectoralmap.files.wordpress.com/2008/04/04-24-race42008-appalchia.jpg



Suggests that in the North Carolina race, which is generally predicted to be an easy win for Obama, he's going to have a real problem in the West of the state, and will have to really rack up the margins in the eastern and central parts of the state...
0 Replies
 
sozobe
 
  1  
Reply Fri 25 Apr, 2008 09:06 am
And West Virginia ain't gonna happen..!

How heavily populated is the western part of NC, do you know?

That could be helpful, if that area -- the mountainous area -- is sparsely populated compared to the rest of the state.

Plus the SC counterparts are nice and blue.
0 Replies
 
nimh
 
  1  
Reply Fri 25 Apr, 2008 09:19 am
nimh wrote:
Suggests that in the North Carolina race, which is generally predicted to be an easy win for Obama, he's going to have a real problem in the West of the state, and will have to really rack up the margins in the eastern and central parts of the state...

Just saw in this article
0 Replies
 
nimh
 
  1  
Reply Fri 25 Apr, 2008 09:31 am
sozobe wrote:
And West Virginia ain't gonna happen..!

Yeah, WV and Kentucky, too, will be horrendous for Obama. Judging on Hillary's margin of victory in the parts of Ohio bordering those two states (see this map of mine from early March), and the results of Tennessee, western Virginia and even southern Illinois, I think anytihing up to a 2:1 victory for her is possible in both states.

So if the Indiana and North Carolina round doesnt finish this off, he's in for a real rollercoaster ride for a bit.

In fact, the article I just linked in (on a pro-Republican blog) built up to a conclusion that Hillary can still win a 100K margin in the popular vote, or even a 500K margin if you count Florida, based largely on the way the upcoming states are positioned in or relative to Appalachia.

That was written back on April 1, and it predicted a 16-point win for Hillary in Pennsylvania. Since she got just 9 points, I think we can take that end prediction with a grain of salt. But the writer was dead on about how the Appalachian portions of Pennsylvania would massively vote for Hillary - see this update.

sozobe wrote:
How heavily populated is the western part of NC, do you know?

Dunno, should look it up..
0 Replies
 
sozobe
 
  1  
Reply Fri 25 Apr, 2008 09:37 am
http://upload.wikimedia.org/wikipedia/commons/1/10/North_Carolina_population_map.png

Not sparse.
0 Replies
 
sozobe
 
  1  
Reply Fri 25 Apr, 2008 09:39 am
Not really dense though either, looks like that's mostly the middle of the state.

Wouldn't the polls take this stuff into consideration? They're showing a big lead for Obama...
0 Replies
 
nimh
 
  1  
Reply Fri 25 Apr, 2008 09:44 am
sozobe wrote:
Wouldn't the polls take this stuff into consideration? They're showing a big lead for Obama...

Yeah they should. Could be that they just have Obama really racking up the vote big time in the central and eastern parts of the state, specifically the "black belt".
0 Replies
 
nimh
 
  1  
Reply Fri 25 Apr, 2008 09:51 am
Hhmm, here's a brand new Survey USA poll of North Carolina (conducted just before the PA primary). It's got a regional breakdown, but it's very broad:

EDIT: here's a SUSA poll done earlier this month - let me include the numbers for both polls below, to reduce the reliance on just one set of data:

Charlotte & West

49% | 48% Clinton
45% ! 44% Obama
6% | 8% Other/Undecided

Raleigh & Greensboro

34% | 35% Clinton
54% | 53% Obama
12% | 12% Other/Undecided

South & Coast

44% | 36% Clinton
44% | 47% Obama
12% | 17% Other/Undecided
0 Replies
 
sozobe
 
  1  
Reply Fri 25 Apr, 2008 09:55 am
Dang, that's closer than I expected. That "Raleigh & Greensboro" section must have a lot of voters, then...?

Charlotte seems to be the red/ orange county in the southwest -- 11 counties over from the western tip, on the southern border. Outside of the Appalachian map, but barely.
0 Replies
 
sozobe
 
  1  
Reply Fri 25 Apr, 2008 09:59 am
Pollster on NC in general (no regions):

http://www.pollster.com/08NCPresDems600.png
0 Replies
 
sozobe
 
  1  
Reply Fri 25 Apr, 2008 10:01 am
This isn't a great map but shows where the cities are -- Raleigh and Greensboro are central:

http://geology.com/state-map/maps/north-carolina-road-map.gif
0 Replies
 
nimh
 
  1  
Reply Fri 25 Apr, 2008 05:58 pm
Uh-oh, that's a little red cross!
0 Replies
 
nimh
 
  1  
Reply Fri 25 Apr, 2008 06:03 pm
Found another recent poll with a regional breakdown: from the John W. Pope Civitus Institute.

This poll has an unusually large number of undecideds: the topline numbers are Obama 44%; Clinton 27%; Not sure 30%.

Re the regional breakdown, keep in mind that the numbers by region become quite small, so you've got to take such specific crosstabs with a grain of salt..

Western

44% Obama
33% Clinton
24% Undecided

unexpected, that..

Charlotte

34% Obama
33% Clinton
34% Undecided

In these two regions Obama beats Clinton by, respectively, 1.3:1 and 1.0x:1. That does make them his least good results in the state.

In the other four regions that are specified (Triad, Triangle, North-East, South-East), he leads by more. By 1.5:1 in the South-East, by about 2:1 in the Triad and Triangle, and by 2.4:1 in the North-East.

Still, nowehere as bad as I'd feared.
0 Replies
 
nimh
 
  1  
Reply Fri 25 Apr, 2008 06:16 pm
nimh wrote:
Found another recent poll with a regional breakdown: from the John W. Pope Civitus Institute.

There's some other interesting stuff in there about the North Carolina primaries too (though the high number of undecideds in the poll reduces its usefulness).

Preference by political philosophy:

The proportions in which voters go to Obama or Clinton is pretty much the same across the board from "Somewhat liberal" to "Very conservative" voters. The only exception are the "Very liberal" voters, who go overwhelmingly for Obama.

The most striking thing here is something else, however. There are about as many respondents here who call themselves conservative as liberal -- and the more conservative the respondent is, the higher the number of undecideds.

The number of undecideds goes up from 16% among "Somewhat liberal" voters to 21% among "Moderates," 39% among "Somewhat conservative" voters, and a whopping 54% among those who are "Very conservative" . And that latter category is still 12% of the respondents. (The respondents are 93% Democrats and 7% unaffiliated.)

Bad news for the general elections, no matter who the candidate? Seems like a lot of conservative Democrats ready to be picked up by McCain...

Preference by voting pattern:

A question about political voter patterns seems to indicate that Hillary is the more electable general elections candidate in this state.

Two-thirds of the respondents said they always voted Dem or voted Dem more often than not. But 17% said they voted for both parties equally and 9% said they voted more often for Republicans. And while Obama beats Hillary about 2:1 among those two-thirds of more or less habitual Dem voters, the floating voters actually prefer Hillary, by a small margin.

That's the opposite of the national pattern of course - normally Hillary does better among core Dems, while Obama does better among Indies and cross-over Democrats. So is it illogical? Keep in mind though: these are not actually Independents or cross-over Republicans; they are Democrats who often vote for Republican candidates. Not the same thing. It could well be that nationally, too, Obama does well among Republicans and Indies who often cross over to the Democrats and badly among Democrats who often cross over to the Republicans - probably very different demographics.

Could it be about race? Blacks make up about 28% of the sample here, and I assume they are overrepresented in the 28% of voters who say they always vote Dem and the 39% who say they vote Dem more often than not. The floating voters here are probably overwhelmingly whites, and Southern whites have shown to be reliably pro-Hillary throughout the primary season.

Indeed, among all whites in the sample, Hillary also edges out Obama, though a full third of them say they're undecided (which in turn probably implies a lack of identification with either candidate, see the bullet point above). Among blacks, on the other hand, Obama leads 7:1.
0 Replies
 
nimh
 
  1  
Reply Fri 25 Apr, 2008 06:19 pm
nimh wrote:
Rasmussen analysis of its daily tracking poll data today: "absolutely no indication that Clinton's victory in Pennsylvania has changed the overall dynamic".

Gallup fiercely disagrees. It has Clinton enjoying a generous post-PA bounce, which is draining the pool of undecideds and bringing her within a point of Obama.

Mind, just like the Rasmussen poll Gallup also has Obama's numbers holding up well: 48% is not bad, right in the middle of a 45-51% band that's he's been in for the past two weeks. But Clinton's number in the Gallup poll is the best she's had since March 21.

Quote:
The latest results, based on Gallup Poll Daily tracking from April 22-24, include two days of interviews conducted entirely after Tuesday's Pennsylvania Democratic primary. Support for Clinton is significantly higher in these post-primary interviews than it was just prior to her Pennsylvania victory, clearly suggesting that Clinton's win there is the catalyst for her increased national support.

Obama's lead dwindled steadily all week, falling from a high of 10 percentage points in interviewing conducted in the three days just prior to the Pennsylvania primary. However, the percentage of Democrats supporting Obama has changed little (declining from 50% in April 19-21 polling to 48% today). Most of Clinton's increased support (from 40% to 47%) has come from previously undecided voters.
0 Replies
 
Eva
 
  1  
Reply Fri 25 Apr, 2008 10:12 pm
How is it possible that there are still ANY undecided voters?

Seems like this thing has been going on forever. Rolling Eyes
0 Replies
 
sozobe
 
  1  
Reply Sat 26 Apr, 2008 06:05 am
I know!
0 Replies
 
sozobe
 
  1  
Reply Sat 26 Apr, 2008 11:46 am
Dunno why my North Carolina map disappeared -- hopefully this will work better. Fits in with the regional patterns we've been discussing:

http://andrewsullivan.theatlantic.com/photos/uncategorized/2008/04/25/results_map33.jpg

I was just reading somewhere (maybe here! if not I can try to find it back on request) that California would most likely have gone for Obama if it had voted later.

Anyway, just from the map, Oregon, South Dakota, Montana and North Carolina seem likely to go for Obama, (and Michigan too, if it had the chance) while Indiana is unclear, West Virginia is unclear (one border for Hillary, one for Obama) and Kentucky is likely to go for Hillary.

West Virginia seems much less unclear in reality (not just going by the map), but I thought it was interesting that everything else lines up pretty well with other indicators.

(My Texas question remains unanswered -- Bill Clinton talked about Hillary winning during the day [primary] and Obama coming in and taking away the victory at night [caucus]. Isn't that what happened? But I digress.)
0 Replies
 
nimh
 
  1  
Reply Sun 27 Apr, 2008 08:40 am
Interesting analysis from Rasmussen in its daily tracking poll update. It can hardly be good business for a daily tracking poll to amphasise how stable and unchanging the race is, but it is what it is:

Quote:
In the race for the Democratic Presidential Nomination, it's Obama 48%, Clinton 42% (see recent Democratic Nomination results). These results are based upon data from a four-day tracking poll. Today's update is the first based entirely upon interviews conducted following the Pennsylvania Primary. On the morning of that Primary, Obama led Clinton 49% to 41%, essentially the same as today's results.

For each of the past six daily updates, Obama's support has stayed between 47% and 49% while Clinton's support has ranged from 41% to 43%. Clinton's support has been within three percentage points of 43% every single day since March 9. Obama's support has been within three percentage points of 48% every day since April 2. During March, Obama's support generally stayed within three percentage points of 46% support.

These results and other data suggest that there is no "momentum" in this Primary Season. Typically, momentum results as voters learn more about a candidate who is doing well. This cycle, Hillary Clinton began the campaign as one of the best known people in the world. Democrats uneasy with her quickly settled on Obama as the chief challenger who has now become the frontrunner. As the candidates have become known, each has developed a solid core of supportive constituencies. For Obama, these included African-Americans, younger voters, more liberal Democrats, and upper-income voters. For Clinton, strength comes from White Women, older voters, more moderate Democrats, and lower-to-middle-income workers.

Six weeks of campaigning in Pennsylvania, and still uncounted millions of dollars in campaign spending, was unable to significantly move any of the demographic support groups from one candidate to the other. At this point, it is hard to imagine anything that will cause the underlying patterns of support to change.
0 Replies
 
nimh
 
  1  
Reply Mon 28 Apr, 2008 08:45 am
Here's a random thought: everyone's discussing how necessary it is for the candidates to target blue-collar voters, working families, regular Joes; to believe the media, you'd think that it's the true proletariat who will decide the outcome of the elections. But the 'proletariat' votes for neither McCain, Obama nor Hillary; they dont vote. We know that, of course, if we think about turnout in general terms, but it's sure something that would slip your mind if you watch or read the immediate election coverage right now.

I was reminded of it when I was looking up census data about Ohio and Pennsylvania, and I was kind of taken aback by how big the difference is between the social make-up of primary voters and that of the population overall.

E.g., much has been made of Hillary's ability to appeal to those with less education, unlike Obama; but in truth neither can.

In Pennsylvania, those with only high school or less make up 56% of the population, according to the 2000 census; but they made up just 28% of the Democratic primary voters. In Ohio, they made up 53% of the population, but just 30% of the Democratic primary voters and 26% of the Republican primary electorate.

Similar with income - although the disbalance is less enormous as with education, it's still imposing. Households with an income of less than $50,000 make up 59.8% of the Ohio population and 60.7% of the Pennsylvania population. But they made up just 44% and 41%, respectively, of the Democratic primary voters in Ohio and Pennsylvania, and just 32% of the Republican voters in Ohio.
0 Replies
 
 

Related Topics

 
Copyright © 2024 MadLab, LLC :: Terms of Service :: Privacy Policy :: Page generated in 0.09 seconds on 11/26/2024 at 02:58:38