Join 3,553 readers in helping fund MetaFilter (Hide)


Barack vs Clinton - a cool interactive graphic
June 8, 2008 8:22 AM   Subscribe

Barack vs Clinton, a cool interactive graphic... What matters when it comes to which candidate people voted for? Race definitely mattered. Income matters sort of. And education. From the NYT, a fine piece of interactive illustration.
posted by storybored (53 comments total) 13 users marked this as a favorite

 
Very cool
posted by fourcheesemac at 8:39 AM on June 8, 2008


With the possible exception of the race category and the flipping of the sex category, these could have been labeled "more liberal candidate" vs "more conservative candidate".
posted by DU at 8:42 AM on June 8, 2008


There's some truth to what DU says; it's fascinating to me when clicking down one of the columns like income, age, or education, to watch the way the whole distribution shifts one way or another.
posted by hincandenza at 8:54 AM on June 8, 2008


Glad you posted this, storybored. It's a very impressive use of technology.
posted by fingers_of_fire at 8:58 AM on June 8, 2008


Extremely well presented info. This should be sold as a piece of interactive software that allows entry of personal information.
Time Spent on Metafilter. Time Spent Having Sex. Time Spent Partying. See yourself as never before.
posted by weapons-grade pandemonium at 8:59 AM on June 8, 2008 [3 favorites]


It would be even better if you could select multiple boxes, e.g., White Men, Age 30-44, No College.
posted by weapons-grade pandemonium at 9:05 AM on June 8, 2008 [5 favorites]


Saw this on the NYT front page last week; very cool indeed.

I think I would have liked for the states to be somehow weighted by population or electoral votes or something, instead of just the raw number of states. Perhaps varying the heights of the boxes?
posted by supercres at 9:13 AM on June 8, 2008 [1 favorite]


Screencast of NY Times graph
posted by jonp72 at 9:13 AM on June 8, 2008


Interesting to go vertically down the rows. That post graduate woman in Arkansas really likes Clinton.
posted by weapons-grade pandemonium at 9:14 AM on June 8, 2008 [1 favorite]


Double?
posted by Brandon Blatcher at 9:21 AM on June 8, 2008


Interesting, too, to follow a single state as you tweak one of the variables. A lot of the time you'll be able to find one or two states that buck a larger trend, which is a nice reminder that voting blocs aren't as monolithic as they look.
posted by nebulawindphone at 9:21 AM on June 8, 2008


This following article is sort of related to this:

The way that polling and demographics slice up the population is, ultimately, a matter of preference; it does not derive from, but is a presupposition of, the "science." Searching for segments of the electorate that vote as a bloc, demographers split the population up into groups they decide are important or salient. And their decisions don't necessarily reflect empirical results -- they are more an index of their own social attitudes, presumptions and prejudices.

Polling's fuzzy math: The idea that a vote turns on one group or another is a deception of demographics by Crispin Sartwell.
posted by Jasper Friendly Bear at 9:26 AM on June 8, 2008


That would be much more useful if the size of the boxes corresponded to the population of the state.
posted by goatdog at 9:49 AM on June 8, 2008


This is one of (the?) most extreme block of voters at 65 points:

"In West Virginia, families earning less than $15,000 favored Clinton by 65 points"
posted by stbalbach at 10:00 AM on June 8, 2008


I saw this and thought about posting it myself yesterday, glad it went up. Very interesting.

Goatdog: I dont' see what difference the state population would make on the graphic, other than more data. Is it necessary or relevant?
posted by absalom at 10:14 AM on June 8, 2008


I am a bit frustrated about the lack of certain caucus states -- like Washington State. Obama scored a dominant win at 68% to Clinton's 31%.

Washington State is below the national average for non-white residents. While one-third of the nation self-identifies as non-white, only 23% in Washington State do the same.

If you sort by whites in the NYT graphic, WA will not appear. It should, and I wonder how many others should? I'm worried about how this racial profiling will be used against the process this November.
posted by andreaazure at 10:22 AM on June 8, 2008


I dont' see what difference the state population would make on the graphic, other than more data. Is it necessary or relevant?

In the graphic as presented, women favored Barack Obama by 19 points in Mississippi and favored Clinton by 18 points in PA. Which do you think adds more "weight" to their respective candidate's total?
posted by supercres at 10:30 AM on June 8, 2008


Interesting, too, to follow a single state as you tweak one of the variables.

I found it interesting that the NYT chose to re-position a state's block, even when it didn't move statistically.

Look at AR & KY in the <30 and 30-44 brackets. The two states don't change side or percentage, but swap vertical places -- for no metric reason I can identify.

I don't know why exactly, but it feels hinky to me. Wouldn't there have been additional programming done specifically to make that happen? If so, why? How many places where a state shouldn't really move, am I being given the impression that there was a lot of data crunching or statistical analysis or movement in a particular demographic, where there simply wasn't?

(I won't be able to hear your replies to my rhetorical questions though, as will be busy at monthly meeting of the MSM Haters Anonymous Club. We're getting new tin foil hats today!)
posted by pineapple at 10:40 AM on June 8, 2008 [2 favorites]


stbalbach: The most extreme group are black illinois voters, favoring Barack at 88%
posted by karuna at 10:53 AM on June 8, 2008


That would be much more useful if the size of the boxes corresponded to the population of the state.

exactly. or if all the states were represented. as it stands, this is misrepresentative of the statistics. you can't just leave off data like that.
posted by mr_book at 10:56 AM on June 8, 2008


Well, you can click all the "swing states" simultaneously if you like. Sure yes, I'd love to see the state's height determined by electoral votes. But you might also color by it's party affiliation. So only tall grey states matter.
posted by jeffburdges at 10:57 AM on June 8, 2008


pinepple: The states are ordered vertically by percentage. Fo r the 30-44 bracket, AR(at 30%) is rightly placed above KY(at 25%). For the <30 bracket, they're both tied at 13%
posted by karuna at 10:58 AM on June 8, 2008


If you listen to techno, then it's fun to click around to the beat. :P
posted by jeffburdges at 10:59 AM on June 8, 2008


I guess I better tell some people back home that there are apparently no black voters in Oklahoma...
posted by Navelgazer at 11:03 AM on June 8, 2008


That said, this goes a log way towards explaining why Obama is choosing to make Georgia into one of his battleground states. Aside from the Bob Barr factor, Georgia cave him the most fervent support across almost all demographics (that this poll measures) of any state in the union.
posted by Navelgazer at 11:08 AM on June 8, 2008


I guess I know who Dr. John Blackguy Jr. of North Carolina voted for.
posted by horsemuth at 11:28 AM on June 8, 2008 [1 favorite]


Tangentially: has there ever been another presidential candidate who was able to market himself equally well by first name or last name?

Something tells me you would never see a post titled "Howard vs. Kerry".
posted by medialyte at 11:34 AM on June 8, 2008


The "Under age 30" button was encouraging.

The "How whites voted" was extremely disturbing.
posted by Avenger at 11:38 AM on June 8, 2008 [5 favorites]


Tangentially: has there ever been another presidential candidate who was able to market himself equally well by first name or last name?

Hillary Clinton?
posted by sbutler at 11:42 AM on June 8, 2008


Dwight Eisenhower?
posted by Justinian at 11:46 AM on June 8, 2008


Maybe I should rephrase that. The "under age 30" category is encouraging because young people are the future, and young people are showing that they don't have a problem with a black president. The "how whites voted" category is extremely disturbing becuse it at least suggests that Obama is going to have a tough time getting white votes in the fall.

It's a generational thing, I guess. Older people are just not thrilled (in general) with the thought of a black President. They voted for Clinton in the primary and will probably vote for McCain in the fall. It depresses me that such a huge chunk of our nation's voters are still trapped in the past.
posted by Avenger at 11:47 AM on June 8, 2008


The most extreme group are black illinois voters, favoring Barack at 88%

Extreme, yes, but very well-spoken of.
posted by Blazecock Pileon at 11:55 AM on June 8, 2008 [1 favorite]


me: Tangentially: has there ever been another presidential candidate who was able to market himself equally well by first name or last name?

sbutler: Hillary Clinton?

See post title. And 15 years of conservatives turning "Hillary" into a slur.

Justinian: Dwight Eisenhower?

Good one. Which makes me wonder if "Barack and Roll" is a campaign slogan that will have a place in history next to "I Like Ike".
posted by medialyte at 1:01 PM on June 8, 2008


Maybe I should rephrase that. The "under age 30" category is encouraging because young people are the future, and young people are showing that they don't have a problem with a black president.

Yes, because the only conceivable reason to support Clinton is because you have a problem with a black president. Conversely, the only reason to support Obama is because you have a problem with women.
posted by moxiedoll at 1:23 PM on June 8, 2008 [1 favorite]


moxiedoll, of course there are individual preferences, but the polling data so far in this election show a strong correlation, in aggregate and given the choice of white woman vs. black man, that people tend to favor a candidate more similar to themselves.
posted by ijoshua at 1:45 PM on June 8, 2008


Also note that states in which the group favored neither candidate are placed in Obama’s column! (e.g. MA and OH Age 30-44; WI women.)
posted by ijoshua at 1:49 PM on June 8, 2008


I should say people tend to favor a candidate whom they perceive to be more similar to themselves, as evidenced in the data on voters’s age and education.
posted by ijoshua at 2:00 PM on June 8, 2008


ijoshua: ...given the choice of white woman vs. black man, that people tend to favor a candidate more similar to themselves.

As a white man, which candidate do I favor?

Or, to put it another way, that was kind of a dumb thing to say.
posted by rusty at 2:01 PM on June 8, 2008 [2 favorites]


...whom they perceive to be...

Ah. And we've diluted it into meaninglessness. That's better. :-)
posted by rusty at 2:02 PM on June 8, 2008


rusty, I already disclaimed that this cannot possibly bear any indication on how an individual will choose (unless you just wanted to lay odds.) The exit polls are available for many states, but I’m not finding demographic breakdowns for ME, so I don’t have enough information to try to wager how you, or someone like you, might’ve voted.
posted by ijoshua at 2:35 PM on June 8, 2008


Could some one tell me if the white voter tended to vote for his whiter side or is blacker side? Or if the black female voter tended to vote for his white side or black side? And did ANYONE vote for him based on the 'information' that his grandmother is white, or that he's a distant cousin of GWB? Did anyone consider the other colors in his blood when voting? How did Obama and Hillary fare among the blind? These are very important questions. I need to know these answers first, before I take a look at his so called 'great' voting record.
posted by Flex1970 at 2:39 PM on June 8, 2008 [1 favorite]


I usually read the NYT in paper form so I'm not that clued into what's happening on their website, but it seems like they have some good staffing for developing useful, interesting interactive graphics like this. I'm thinking about the rent v buy calculator from a while ago, and I think I saw a delegate count tool where you moved sliders to play out different assumptions for how the primary end game could go.
posted by yarrow at 2:44 PM on June 8, 2008


The "Under age 30" button was encouraging.

The "How whites voted" was extremely disturbing.
posted by Avenger at 2:38 PM on June 8 [3 favorites +] [!]

I saw a lot of white voters crossing over to vote for Obama but not the other way around. I also saw men shifting over to Obama and away from Clinton. So the question remains for me, will those men who voted against Hillary vote for Obama in the fall? Or will they go over to McCain because they couldn't stand Clinton and wanted to knock her off in theprimary?
posted by etaoin at 2:57 PM on June 8, 2008


Question--what language was this little app written in? I know it's flash, but what language was used to write it?
posted by zardoz at 4:37 PM on June 8, 2008


Actionscript, I would hqve thought - it's a real language these days.
posted by Artw at 4:50 PM on June 8, 2008 [1 favorite]


It would be even better if you could select multiple boxes, e.g., White Men, Age 30-44, No College.

why do you want to blow up your monitor?
posted by matteo at 6:05 PM on June 8, 2008


Not much difference between undergrad and postgrad.
posted by Afroblanco at 7:08 PM on June 8, 2008


Apparently nobody under 30 votes in Oklahoma, which is consistent with my experience.
posted by yhbc at 7:22 PM on June 8, 2008


I just thought I'd mention that one of the ads on this page (for those of you who aren't logged in) is a fake poll which asks "Is it OK to unconditionally meet with anti-American foreign leaders?", with a picture of Obama and Ahmadinejad side-by-side. Paid for by McCain 2008.

It's gonna be a long one, folks.
posted by Avenger at 8:06 PM on June 8, 2008


with a picture of Obama and Ahmadinejad side-by-side

Obama has already lost the Hamas endorsement, so I'm not sure how long he can hang on to Ahmadinejad. At this point even Kim Jong-il's vote is up for grabs.
posted by lukemeister at 11:09 PM on June 8, 2008


"Is it OK to unconditionally meet with anti-American foreign leaders?"

Um, yes. All terrorist/freedom fighter conflicts that have ever been resolved have been done so by talking first, cessation of violence second. Anti-american foreign leaders are the MOST important people to be talking to if you want to make then into pro-americans or at least neutrals.

Sorry, was that a trick question?
posted by ArkhanJG at 11:51 PM on June 8, 2008


"Is it OK to unconditionally meet with anti-American foreign leaders?"

Of course not. Clearly, everyone should agree with us before we talk about anything.

WTF.
posted by rokusan at 2:04 AM on June 9, 2008


Very nice, easy to understand visual depiction of the voting breakdown. Interesting to see where the dividing line fell.
posted by geeky at 10:53 AM on June 9, 2008


« Older you've got new postcrossed mail...  |  Special Containment Procedures... Newer »


This thread has been archived and is closed to new comments