
Here's a neat map of the primary results in the Northeast that lets you see trends that can get obscured by geographical borders.
« The Bukharin Factor | Main | The Next McGovern? » Votes Without Borders24 Apr 2008 08:24 am
Here's a neat map of the primary results in the Northeast that lets you see trends that can get obscured by geographical borders. Comments (29)
That Bukarin thread really demonstrates what weird criteria you seem to have for what threads to weigh in on and which comments to respond to. Not that I'm complaining, you can talk about whatever you want to. Just think it's interesting.
The national county-by-county map is even more striking. It gives you a real sense of how Clinton's suppport is highly concentrated in the interior South and Appalachia, just where one would expect Bill Clinton (as opposed to Hillary Clinton if she wasn't associated with Bill Clinton) to be strongest. And if you mapped just white non-Hispanic voters, you would probably get more of the South and Northeast Corridor for Clinton, but her support in the Southwest would be worse (thanks to excluding Hispanic voters). And that is why she is losing. She is actually just a regional candidate, largely dependent on people who feel a very strong affinity for Bill Clinton plus the Democratic machines controlled by people loyal to the Clintons. In contrast, Obama gets support from every region not strongly tied to Bill Clinton or dominated by Clinton-controlled machines, and that is why he is winning.
I followed the link, but they didn't have an easy to read guide on which counties count and which counties don't count. Should I just guess based on the color coded results?
Is the stronger correlation education level, gender or population density? And, what is the correlation between education level and population density? Clinton does best in rural (low population density) counties, does best with the less educated, and does best with women. Which of these is the most telling?
Freddie:
Hey, as long as it isn't ruminations on OMIGOD tha Wire iz the best EVAR! then I'm okay with it.
Doug, Really, none of the above. If you get outside Clinton's core regions, Obama starts doing much better among all three of your categories. So, the strongest correlation is with region, not any demographic factor.
Definitely illustrates Obama's college town advantage. Can anyone guess where Cornell is on the NY map?
Definitely illustrates Obama's college town advantage. Can anyone guess where Cornell is on the NY map? And where Penn State University is.
Deibs, pretending for just a minute that the Clinton camp is acting in good faith rather than grasping at straws, there is a difference, I suppose. NY and PA are pretty lightly populated along both sides of the border (although the Buffalo and Elmira media markets do eek into PA just a bit). In contrast, northwest Indiana is pretty densely populated. About 10 percent of Indiana's population is within the Chicago media market, but given the demographics of that area (lots of African-Americans, Hispanics, ethnic Catholics, and eastern Europeans) it's estimated that something like 25 percent of Indiana's Democrats live in the Chicago media market, and therefore those who watch local news have been familiar with Obama for much longer than the typical non-Illinoisan. There also are a good number of transplanted Illinoisans in "the Region," as it is known, and they may have some interest in seeing someon from their native state in the White House. I'm still skeptical about the polls that show Obama ahead, because so many of the state's Democrats are demographically similar to Democrats in Ohio and PA (although there are just fewer Democrats here). Plus, Senator Bayh, now only my list of Democrats I will never lift a finger for again, is putting himself on the line for Hillary Quixote. But Obama should have a bit of an advantage in that part of the state.
So what's witht the difference between western Mass and Connecticut? That seems to be pretty starkly divided by the border, and CT is more in the NY media market than Mass, which would mostly get Springfield tv, I think.
The most striking thing to me is the deep blue belt that covers the Appalachians. If you could see all of Virgina on this map you'd see that that was where Obama hit the reef (in a state he won by a landslide), as soon as the vote got out to whites in the mountain regions. Clinton's best margins in PA came out of the Lehigh valley (where she has family connections, as voters were told again and again) and the Appalachians. The lack of Appalachian whites, the growing population and economy, and the lower proportion of Catholics in Indiana are going to make it a lot tougher to build the Clinton coalition, which is probably why Bill Clinton is running around telling the relatively affluent working class whites that Obama hates them anyway.
Interesting commentary on the value of endorsements - isn't the PA picture pretty much a negative of what you would expect if there was any value to Casey's support of Obama and Rendell's of Clinton?
But Dave, that's where it gets tricky when considering a primary as opposed to a general election. Compared to Illinois, Michigan, and Ohio, Indiana is less urban/more rural, less white, and less ethnic. But it's also less Democratic. So, while there are fewer Catholics as a percentage of the population, I'm not sure that the demographics of the (smaller) Democratic coalition in Indiana are that much different from the larger coalitions in surrounding states. What's interesting about Indiana is that it has been Republican for as long as there have been Republicans. So, unlike some southern/border states such as Missouri and Kentucky, there's no long tradition of Democratic Party affiliation among white protestants. Indiana does have a bit of an Appalachian streak (geography aside), but that part of the population isn't necessarily Democratic. I'm really excited and interested to see how this turns out. Between the late primary and the reliably Republican outcome in presidential elections, there just has never been much attention paid to Indiana's presidential politics. I really have no idea how it's going to turn out.
Obviously I meant Indiana is more white, not less white.
The national county-by-county map is even more striking. Yep. You look at the national map and see the missing pieces of West Virginia and Kentucky and can tell just from the map her strength there. For North Carolina, you can see Virginia above and South Carolina below and see that Obama should take everything east of Asheville.
Looks to me that Obama is strongest in cities (look at Essex and Mercer counties (Newark and Trenton) in NJ, Philadelphia and Baltimore and DC) and majority-black areas such PG county in MD (can't talk about race though - otherwise the Obama-bots get after you as being a "racist" the way they attacked Ferraro for stating the obvious)- once you get into less urban/more white areas Hillary tends to rule. I'd say when it comes to electability and holding onto and nergizing the Democratic base, the problem at this point is more Hillary's than it is Barak's.
I am personally predicting a small win for Obama in Indiana. The southern part will likely shade into Clinton's South/Appalachian territory, and the northwest part is Chicago suburbs and likely Obama territory. The central and northeast parts will likely be similar to the neighboring parts of non-Appalachia Ohio, with Obama doing relatively well in cities like South Bend, Indianapolis, and Fort Wayne, just as he did in Columbus and Dayton. And I think if you add that all up, Obama gets the advantage based on superior numbers in his territories as opposed to Clinton's territories. But I think the southern part of Indiana will keep it considerably closer than, say, Wisconsin.
Ethel-to-Tilly, It is not that we can't talk about race, but I think we need to look at the whole map first (available through the link above). Again, once you get out of the South and west of Appalachia, all those demographic trends fall apart. For example, in those regions Obama is doing very well in largely white rural areas. Indeed, I suspect there are demographically similar counties in different regions which approach a 40-50 point swing. So, race alone fails to explain what we are seeing. Indeed, all the conventional demographic factors together fail to explain what we are seeing. Rather, the dominant factor is regionalism, as the national map demonstrates handily.
mark f, I suspect that is largely a party machine effect (my sense is that Massachusetts is much more a machine state than Connecticut).
If you follow the links, there's a map for Democrats abroad. So, what I want to know is, why doesn't anybody mention the Clinton strongholds of Libya, Liberia, Somalia, Yemen, and Turkmenistan? If a Democrat can't win there, he can't win the general!
Here's a weird trend. Below is a list of Obama's performance among white voters by region: If you adjust for education Massachusetts and New Jersey are in the same category as 5. I don't really have an explanation for why Obama does that much better among whites in the Southwest and the bigger states in the south.
I suspect that is largely a party machine effect (my sense is that Massachusetts is much more a machine state than Connecticut). - DTM Indeed. Machine dominated Middlesex county seems to buck the trend of Obama to dominate the I-95 corridor. What's interesting to me is how much Obama does well in Democratic regions. One could argue that the geographic distribution of votes shows HRC has a better chance of capturing swing voters, so, in the end, HRC has a better chance of winning. OTOH, one could argue that elections nowadays are about getting the base out and Obama does well where people actually do go out and vote Democratic, so we'd be more likely to win with Obama. But what's even more interesting to me is that for all the spin from the Obama campaign about transcending ideological divisions, he seems to mainly appeal to liberals. And for all the talk you hear from religious (especially Catholic) conservatives who claim to be moderate or even liberal on economic and national security issues (and hence you'd think they'd support Obama over the hawkish, economically neo-liberal HRC ... as both have about the same positions on social issues and hence would be equally "bad" for social conservatives on social issues while Obama would be "better" on economic/national security issues), HRC seems to be appealing more to religious conservatives than Obama. Is this an issue of race? Or perhaps, for all their rhetoric about their religiously derived political views transcending the political spectrum, religious (especially Catholic) conservatives really are politically conservative? Is the political spectrum really more 1-D than some social conservatives who claim they'd vote for the right kind of Democrat actually claim? And what does that mean for the likelihood of HRC winning because of swing votes? Is she, being more moderate, more likely to win, in spite of those swing voters voicing opinions actually closer to Obama? Or is she too much getting into the right and simply will alienate the base whilst people who would prefer her to Obama would prefer McCain to her -- so they'd vote for McCain. Certainly, too many of her anti-Obama ads could be turned around and used against her by McCain -- which shows the degree of her, um, political "wisdom". But Obama called bitter people bitter ... and we can't have that ...
You're right DTM - I see what you mean when you look at the larger map
Boy, people in Idaho really don't like the Clintons.
Robert, I think the answer to your question is pretty obvious once you flip it around to look at how Clinton is doing among white voters. Specifically, suppose you took a candidate who was perceived as essentially one part Northeast Corridor, and one part Arkansas (through a proxy effect for Bill Clinton). Now flip your list around. Isn't that pretty much exactly what you would expect to see for such a person? And that I think is the basic problem with so much of the coverage of the contest. People are looking for reasons why certain white people are voting AGAINST Obama, and aren't coming up with good explanations for the patterns we are observing. But if you flip it around and ask why those same white people are voting FOR Clinton, it all makes sense.
DAS, Last I saw, when you sum up all the available exit polls, nationally Obama was solidly beating Clinton among both self-described liberals and self-described conservatives, with Clinton keeping it close only among self-described moderates. What that suggests is that Obama does indeed appeal to a broader ideological spectrum, and conversely that Clinton only appeals to those who are aligned relatively closely with her ideologically. Indeed, that distribution suggests that ideology is only one of the relevant dimensions determining voter preference, and that when summed up, Obama is beating Clinton solidly on the dimensions orthogonal to ideology.
For North Carolina, you can see Virginia above and South Carolina below and see that Obama should take everything east of Asheville. Well, let's say east of US 321: Gastonia-Hickory-Boone. Hillary was in Asheville last night, and Bill was there a few weeks ago. The west of NC is one of those places where 'white blue-collar' voters -- i.e. men with baseball caps and moustaches -- are probably registered Republicans.
Comments closed May 08, 2008. |
Copyright © 2008 by The Atlantic Monthly Group. All rights reserved.
We're hearing comments pointing out Clinton is at a disadvantage in Indiana because it borders Obama's home state. Why didn't we hear so much about Pennsylvania's border with New York?
Posted by deibs | April 24, 2008 8:56 AM