For the most of you who haven't followed my baseball work, I am best known for inventing a forecasting system called PECOTA, which generates predictions by comparing baseball players with a large database of historical peers and identifying the most similar ones. This same technology -- which is really just a variant of nearest neighbor analysis -- can be applied to virtually anything, including identifying the similarity of any two states along a number of dimensions of political salience. In fact, that's exactly what I've done in the chart below, with each state listed along with its three most similar states.
What factors go into the similarity score? There are quite a few, which are weighted in rough proportion to their importance in determining the Kerry-Bush result in 2004 and the McCain-Obama polling this year according to an analysis of variance.
Specifically, those variables are: (1) Partisan ID index; (2) Likert liberal-conservative score; (3) Average years of completed schooling per adult; (4) Per Capita Income; (5) 18-29 year old population; (6) senior population; (7) African-American population; (8) Hispanic population; (9) percentage of white evangelicals; (10) Catholic population; (11) Mormon/LDS population; (12) percentage of military veterans; (13) percentage of same-sex partner households; (14) gun ownership rate; (15) percentage of adults identifying ancestry as 'American'; (16) percentage suburban; (17) percentage of state jobs in manufacturing sector; (18) current unemployment rate, and (19) latitude and longitude (e.g. geographic distance).
The highest score theoretically achievable is 100, for two states that are exactly identical along each of these 19 dimensions. The highest score in practice is 71 between North and South Carolina. A score of 0 represents states that are as dissimilar as similar, and negative scores are both possible and quite common (though I list them as zeroes in the table above).
Note that some states really aren't like any other states at all, including big ones like Florida and Texas and small ones like Alaska, Utah, and New Mexico. Then there are other states that are sort of within the main sequence but need to pull from different regions -- like Indiana, whose three most similar states span the Midwest (Ohio), South (North Carolina) and the Prairies (Kansas).
And yes, this does have implications for our model, which will become clear at some point soon.
7.07.2008
State Similarity Scores
@ 8:18 AM
Subscribe to:
Post Comments (Atom)
87 comments
So states like Utah and (DC) are so unique that nothing is anything like them? and HI is almost like that, but if i wanted to know at least with 10% accuracy what HI is like i ought to go to CO?
hm... interesting. i wonder how we could visualize this in a compelling way.
As far as I can see, all of the states with similarity scores greater than 60 are projected to go the same way, with one exception: Missouri. It's closest neighbor is Ohio, but Ohio is projected for McCain, and MO for Obama. I wonder if the old saying "as Missouri goes, so goes the nation" will be wrong in 2008. Anyone know the last time MO and OH went for different candidates in a presidential election?
Oops. Above I meant OH for Obama, and MO for McCain.
Florida's set is peculiar. I would have had a hard time guessing one of those states as part of its group.
None of Florida's comparables are any good. Pennsylvania and Arizona show up because they have a lot of old people (Arizona also has a lot of Hispanics, though diverges from FL politically). I'm not really sure what Delaware's doing there.
Pretty cool, Nate did these have predictive value in past elections to a significant degree? It may prove insightful to predict outcomes based on state patterns. Look forward to your manipulation of it.
I love this, Nate. One question: is there any data on percentage of people born in the state v. transplants? I'm very interested in this particular set of people (being one of them, to the seventh degree).
I'm assuming that people will see a connection between education level and inter-state moves, but it's more than that. Living in Texas, you could even say that an intra-state move voters are more similar than those who have stayed within their region.
Just wondering if that's where you'd find the difference that are so obvious to those who live in North and South Carolina. (How could they possibly have edged out the Dakotas?)
Nate, do you think you can figure out what sort of similarity scores are significantly similar (or different)? I know that with the Jamesian creation, similarity scores of over 950 are very significant, 900 are "pretty" significant, and ones under 850 are not really that significant at all. Just like when you have Barry Bonds's best comparable to be Willie Mays at around 760, that means that Barry is in a league of his own...
What similarity score puts something in a league of it's own? Obviously, Alaska and Hawaii have their own eccentricities, and the Carolinas and the Dakotas swing pretty together - but where's the (fuzzy) line in between them?
I know I'm asking more questions than posing answers, but I hope that they're at least good questions! =)
Nate said "None of Florida's comparables are any good. Pennsylvania and Arizona show up because they have a lot of old people (Arizona also has a lot of Hispanics, though diverges from FL politically). I'm not really sure what Delaware's doing there."
I can offer some reasons why Delaware may be a "nearest neighbor" to Florida.
The party IDs of both states are probably very similar. The Wilmington/Newark area has a relatively high percentage of Republican identification, though many of the Republicans in this metropolitan area are more in the "Rockefeller Republican" mold, and are much more liberal on social issues.
Given the fierce identification of many Delaware Republicans with the party (despite the disparity in beliefs about social issues), the Likert scores for Delaware are probably over-ranked towards Republican identification.
Also, there is a tremendous disparity in Delaware between the Wilmington/Newark area and "downstate." Wilmington/Newark is probably similar in many respects to south Florida (other than average age), while downstate is comparable to northern Forida and the panhandle.
All in all, there are some good reasons why Delaware might be identified as a nearest neighbor to Florida, but it is a conclusion which will not permit any further inferences to be drawn, since Delaware's score is very distant from Florida's score, especially for being a nearest neighbor.
As someone who was a reader at Baseball Prospectus long before FiveThirtyEight existed, let me state for the record that one person being responsible for the creation of either PECOTA or 538 would be truly impressive; that one person is responsible for both is nothing short of remarkable.
And yes, Nate, this new tool feels quite familiar. When can we expect a post evaluating McCain's potential VP picks by Popularity Above Replacement Governor?
Nate,
Could you put a 'national' score on this chart? I've heard several folks say that a Florida election is like a mini-national election. It might be nice to know which state looks most like the national average.
Once again, very cool. Another great way of quantifying conventional wisdom.
I have a few questions: (surprise!)
1) I'm not sure I understand why you have introduced the discontinuity by forbidding negative numbers. By setting all negative numbers to zero, are you effectively saying that DC is as similar to Maryland as it is to Utah?
2) In a nearest-neighbor analysis, where do the negative numbers come from? Any sort of measure of distance should give positive numbers, and it should not be too hard to translate distance into a scale of (1 = identical, 0 = infinitely far apart).
3) How do you determine that states are "as dissimilar as they are similar?" This makes sense in the context of a correlation-like metric, but such a metric is dependent on the sample mean. Does this mean that in your metric, the similarity between two states may change as a result of changes in the demographics of the other states?
Off-topic
I couldn't help but think of this site when I saw this map on Sullivan. Is Obama's real weakness among fat white people? Perhaps he should pack on the pounds. :-)
Hi Nate,
This is fascinating!
Seeing your chart, the gears in my mind start turning on a dozen fronts, and it sounds to me like your state similarity scores could have a bearing on a million other applications -
When it's ready, will you consider representing the data geographically, instead of in table form? I think a color-coded or shaded/graded US state map highlighting these relationships would be very enlightening and useful.
Secondly, what is your stance on sharing these kinds of datasets themselves? I start drooling when I think of a spreadsheet with (your) "Likert liberal-conservative scale" by state (something I've dug around for on the net and have come up dry). I know a lot of this data is publicly available or in the census, but it is damn hard to track down. I'm doing some GIS work related to state progressive politics and environmental policy, and this is the kind of thing that could be very helpful.
Have you thought about making certain sets of data available to the public through fivethirtyeight? If not, is there anything you'd be willing to share?
Either way, hope to see a state-similarity map sometime soon!
-Jeremy
Link
Here's another place where your analytical & baseball expertise can help answer a question.
Nate,
your data isn't symmetric! For example, in the row labeled "Idaho" you have "Wyoming 47"; in the row labeled "Wyoming" you have "Idaho 40".
But nice work! It's possible to derive a sort of "map" of the states from this, which I've made a first stab at and which is interesting in ways I'm having trouble articulating; I'll probably post it to my own blog soon.
According to the secretary of state's office, 55,560 more Democrats than Republicans are on the active voter rolls in Nevada, as of the end of June. The gap widened from 50,020 in May and represents 5 percent of the 1,031,984 active voters.
http://www.lvrj.com/news/24004419.html
With a big tournout in Clark County, Obama will win in Nevada.
Kerry lost by only 21.000 votes and the Republicans had a 1 percent edge in voter registration in 2004.
I think you'd be better off having separate axes for percentage of adults with various education levels: completing high school, some college or associates, completing bachelors, completing a graduate or professional degree. There's a difference between a state where 40% of the population has gone to community college and 15% has gone to a 4-year college (Wisconsin?) and one where 30% has gone to college and 10% has gone to community college (Virginia?). In your model they would end up the same.
Ok - you need to create a page where we select a state and every other state gets a "heat chart" index for the selected state.
Using this we can visually see how similar a state is to others.
As a little old librarian/web designer/cartographer-wannabe, I recommend either a key or a link to an actual map to make the color squares more meaningful. Frankly, the squares of color are useful from a separating-the-columns point of view, but I'm not sure they'd pass the Tufte test. Then again, this site is dedicated to data, not design. (But is it possible to separate the two at this point?)
Are your regions presented somewhere else? More than likely that I missed something somewhere.
(Finally, a minor gripe: including Connecticut in the New England block seems more nostalgic than realistic. Any comments?)
The most immediate reaction I had to this analysis was, "Ahhh, so that's why Obama has put South Carolina on his map today!" He's in Charlotte, NC today, but has declared his intention to campaign for SC along with NC and GA.
Before your analysis, I would have thought SC would be too much of a reach - NC "feels" closer to VA than to SC, at least in the urban centers that contain most of the votes. But I have also thought that the whole South Atlantic corridor is a possible Obama pickup - and certainly good strategy for the Dems longterm, since this is where the country is growing, and "bluing."
Interesting.
Prematurely Grey,
I agree with your comments about the colors, but I have to say that in general Nate does a pretty good job visually presenting data. The regions are presented in the middle of the left-most column of the home page.
As for CT, it certainly seems New-England-ish to me. I think the rural/suburban parts of NY, NJ, and CT are similar to the rural/suburban parts of New England. The difference is that outside of MA, there isn't really a major urban center in New England. That's probably why Nate's analysis puts CT so close to NJ and MA. Boston affects MA similarly to how NY affects NJ and CT.
In a nearest-neighbor analysis, where do the negative numbers come from? Any sort of measure of distance should give positive numbers, and it should not be too hard to translate distance into a scale of (1 = identical, 0 = infinitely far apart).
Yup! In this case, just take x to 2^((x-100)/100) and you're there!
Anyway, from where I'm sitting, these aren't distance measures per se, they're more like a rough measure of the correlation/covariance between two states. (The "portfolio theory" tag would seem to hint at that...) And as we know, indices of correlation can be and often are negative.
Although if that's the case, why isn't there a variable for historical correlation between two states' votes? Maybe it's just that that would require 2500 variables rather than 50, but even so, what about the "bellwether value" of a state, what in finance would correspond to a state's beta coefficient?
As others have noted, fascinating, Nate. My work takes me from Washington state to a number of other states around the country working with county governments. The similarity to Oregon is not surprising, but it was gratifying to find that Minnesota was also relatively similar to Washington, an antecdotal observation I've made in the past.
Having done some similar work many years ago in comparing state political cultures using factor analysis, (as I said, it was many years ago), the problem of the "big" states like Florida, Texas, New York, and California is familiar to me. Simply put, heterogeneous states are, not surprisingly, more difficult to classify and size is strongly correlated with heterogeneity.
The last time Ohio and Missouri went for different candidates was in 1960.
Utterly fascinating, as usual, Nate. I notice that both New Hampshire and Illinois list Minnesota as their nearest state, while Minnesota has neither Illinois nor New Hampshire in its top three. Ohio is listed four times in the second column, and there are three states listed three times in the second column: Minnesota, Colorado, and Alabama. Perhaps these are "archetypal" states that a lot of other states are close to?
Nate, I think you're onto something with this analysis. The more I look at it the more it accords with intuitive common sense.
For instance, notice that the most homogeneous states are the deep South:
North Carolina is 71% like South Carolina
Georgia is 62% like South Carolina
Alabama is 56% like Tennessee
Tennessee is 68% like Kentucky
They all link together.
The Midwest works too:
Kansas is 55% like Nebraska
Nebraska is 61% like South Dakota
South Dakota is 70% like North Dakota and vice-versa.
Massachusetts is only 52% like Connecticut. Growing up in Connecticut, culturally the eastern 2/3 of Connecticut is identical to Mass (they are Red Socks fans, not Yankees fans for instance). But, voting wise, it is less liberal. This analysis points out some demographic reasons why that would be true.
Notice that Rhode Island is closer to New York (only 49%) than either Mass. or Connecticut, despite being squeezed between them.
Harrison,
I'm not sure how your distance scaling function works. If I'm reading it correctly, it would take a distance of 0 and assign a score of 1/2. Is that right?
Nate used to have a list of states labeled "must-win" states. Several of the commenters (myself included) thought that the list corresponded more to "bellwether" or "telltale" states, i.e. states for which you could reasonably say "As goes X, so goes the nation." I think the best metric of such states would be correlation between how the state voted and who won the election in Nate's simulations. Maybe some day Nate will bring the list back in fancy graphical form, as he did for tipping-point states.
The biggest difference between North Carolina and South Carolina is a historical/cultural one not reflected in Nate's data. I would call it the Confederate factor.
SC arguably was the seat of slavery in America, it started the Civil War almost single handedly, then it was on the receiving end of much of Sherman's March to the Sea. When the Civil Rights movement started, SC started flying a confederate flag above its statehouse.
In contrast, only half of NC was really slavery country, so it was one of the last states to secede (6 months after SC),
and the vote to secede passed the NC legislature by one vote, and yet, NC went on to lose 3 times as many troops as SC did. When the Civil Rights movement started, NC kept its own flag, and didn't put up the confederate flag.
These different histories are still reflected in aspects of the states' cultures - you see a lot less confederate flags in NC than you do in SC, and in general, blacks and Northerners often feel more welcome in NC.
Interesting that this post is tagged with "Indiana" - are you thinking, Nate, that as goes Ohio, so goes Indiana as well?
Now we're getting somewhere very interesting!
I'd though drop out variables (5),(6) and (13)(this is as about as relevant as percent of people who live on farms).
Instead I'd add "PI[i]GS" variable from the Ancestry question: P_olish, I_talian, Irish, G_reeks, S_lovak, and some others (Portugeese, Hungarian, Lithianian, Arab as well as some others to account for West Coast Ethnicity but that messes up the late Msgr. Baroni's acronym). The "Non-WASP Ethnic" is a far more relevant variable. It's not quite the inverse of "american/US ancestry) because it sifts out the West of the Mississippi but East of Pacific Coast, while showing why CO, NV are different and why VA is different from the South.
Regarding my comment about the Confederate factor, I should have clarified that I think that factor explains why NC and SC have similar conservative scores, yet people often consider SC more conservative than NC. SC's outward appearance is often more conservative, even if the two states have similar scores.
Also, I should clarify that I am not claiming that NC's western half, which was reluctant to secede, was liberal or abolishist - most of that sentiment was more about not wanting to fight someone else's war and/or wanting to remain part of the Union.
This approach is interesting. I wonder if you can use PECOTE to estimate what baseball players are using performance enhancing drugs.
Gelbstoff
This is really fascinating -- beyond the election, just as a measure of what states are close to each other on a cultural level. Based on the states I lived in, it feels very accurate.
Could you release the full charts, not just the top three? I'd be interested in seeing the others.
Have you thought about representing this visually, perhaps drawing lines between states with a score of 50 or above and seeing what kind of cloud clusters result?
It would be hard to do on a map. Some of the clusters are geographically contained, like South Carolina-North Carolina-Georgia. But others are more interesting, like Washington-Oregon-Minnesota-Colorado. That might be hard to represent on a standard map, but some kind of rearranged map might work.
One strange thing is that Florida and Delaware are supposedly somewhat similar in your model and yet vote very different. I'm not too sure how your model works, but Florida is unique probably because of its large Hispanic AND African-American population. There's also a large number of moxed-race Hispanic blacks. Did you take that into account?
I'm not sure how your model works, but have you thought about overlapping variables and how best to avoid those? I particularly dislike "latitude and longitude". You probably included the variable because closer states tend to vote together, but this is in fact reflected in many other variables and distance itself has no direct effect. It becomes inaccurate, especially, after you take into account population density and other factors. Idaho and North Dakota, for example, are quite far but not too different ideologically because they're very rural and thus the "cultural distance" between them does not as much in relation to actual distance. Of course, being rural itself tends to have a relationship with being conservative. Another example - California and New York are at the two ends of the U.S. and yet they're similar ideologically - probably because they're both coast states, which brings another point. Being near the coast has a little actual effect on ideological differences, but it probably brings a higher population density and this has an effect on the ideological differences as well, which is reflected in your 16th variable, percentage suburban. Or another example of why distance itself often doesn't work - Delaware and Florida tends to vote more closely than, say, Delaware and West Virginia, which is closer. This could be because West Virginia is mountainous and thus it is easier to go to Florida from Delaware than to West Virginia. This mountainous variable is also related to black population - it's difficult, after all, to create plantations on mountains.
All variables overlap somehow, but the latitude and longitude variable seems only to bring close states closer and vice versa by overlapping other variables, making it useless. It's also hard to measure when you take into account the population density, etc. I'm sure you found it useful or you wouldn't have used it. I'm just puzzled as to why.
Modeler: Absolutely. Think of it as taking Nate's values and transforming them into the probability that two states will vote together; if two states are uncorrelated, then Nate will give them a value of 0, and they have about a 50% chance of voting together.
I've made a simple attempt to try to visualize this data. See my blog for a brief explanation (basically, I use the similarity scores as weights in a minimum-weight spanning tree, if that means anything to you) and see here for the picture.
harrison,
I don't think Nate's similarity scores can be interpreted quite so directly as probabilities.
In reference to Isabel Lugo...
I love this site. It has to have the greatest signal to noise ratio of any site on the net.
Thanks, Isabel.
"Could you put a 'national' score on this chart? I've heard several folks say that a Florida election is like a mini-national election. It might be nice to know which state looks most like the national average."
I think he made a post to this effect a little while ago that had Nevada as the closest state to the national vote.
anonymous @7:53 AM wrote:
"I wonder if the old saying 'as Missouri goes, so goes the nation' will be wrong in 2008"
The actual old saying is "As Maine goes, so goes the nation".
http://en.wikipedia.org/wiki/As_Maine_goes,_so_goes_the_country
The "Missouri bellweather" idea is much more recent.
Wow. Fascinating. I can see how this would be very useful in your projections. Truth be told, I'd have to look at it for a while to really understand it, and get a lot more info on the major demographic similarities and differences to get a handle on it...But it's clear that you have a grasp of it. No need to do more work turning it into a viewer-friendly graph/chart; as long as it works for your uses.
Your model is becoming quite comprehensive. The one major factor that seems to be missing, in my mind, is an estimation of organizational strength. Not only of the Obama and McCain campaigns, which will have room to maneuver in the next four months, or traditional (or projected rising/waning) Democratic and Republican Party organizational effectiveness...But certain "interest groups." (I'm using a broad definition, from the NRA to the Catholic church.) Interest groups can exercise disproportional strength to the local electorate. For example, (excuse my laziness in not pulling statistics) most Rhode Islanders favor Roe vs. Wade--keeping abortion "safe, legal, and rare." But, laws currently trumped by Roe are on the books, which would severely limit abortions in-state. Unlike many other states, RI lawmakers are unwilling to pass a law keeping anti-abortion protesters at least x feet from clinic property. Clinics that perform abortions in RI are constantly disrupted by angry protesters as a result. The Catholic churches frequently hold pro-life events for several hundred people, while pro-choice events can't seem to gather more than a few dozen--despite (or due to, depending on your perspective) a majority of Rhode Islanders supporting current federal law.
That particular example isn't important for Presidential politics, but it controls state politics. I'm sure there are similar cases across the country. It is difficult to measure, with "voter enthusiasm," "wedge issues," and specific (or, lack of traditional/expected) endorsements/speeches/sermons playing a role.
---
Modeler (on your off-topic): I know your statement was probably tongue-in-cheek, but...Just for fun, here's a map! http://i191.photobucket.com/albums/z53/MDA2007/calstates.jpg Of course, correlation != causation.
Firstly, this is interesting work. Nate can you retrofit your similarity scores to the actual results of the Hillary-Obama primary?
This analysis is informative, but not predictive.
Ohio and Michigan have the one of the highest matches at 66 yet they often voted differently.
Reason is Rove's micro-targeting of church goers in Ohio.
Given that the organizational advantage is Obama's this year I think we can expect CO, VA, NM, OH, MI to be bluer than this would predict.
Isabel: You're probably right in that any attempt to interpret them as probabilities will be (a) highly complex, (b) highly wrong, or (c) both. But if you want to look at them that way (and I do, since that's how I personally tend to look at things), then I'd think that a similarity score of 1 would correspond to a 100% chance of two states voting together, a 0 would correspond to a 50% chance, etc.
Here's my guess as a Silverologist.
Nate has discussed recently the observation that
GA smaller than NC smaller than VA
Making the point that this is not just a Kerry-to-Obama swing relationship: it has to do with the specific features of the States. The kind of strategy that will lead you to a capture of GA would already have led you to capturing NC and VA.
I.e., instead of envisaging a single domino of all Bush States arranged by a required Kerry-to-Obama swing, we should envisage separate Paths of Conquest:
Electoral Strategy A brings us first State A1, then State A2, then State A3 etc. (similar states arranged by their required Kerry-to-Obama swings, adjusted for the shared demographics).
Electoral Strategy B brings us first State B1, then State B2, then State B3, etc.
Note the beauty of the exercise: there is no politics involved. We do not work this out from "which campaign should appeal to which demographics?" Rather, we let the data speak and find that there OUGHT to be a campaign strategy that appeals to States A1, A2 and A3., that there OUGHT to be another campaign strategy that appeals to States B1, B2 and B3, etc. Once we see the various lists of States, then our political strategist can figure out just what is the campaign that does appeal to them; but this is not required for finding that such a campaign exists.
Once we have figured all of that out, we can see how the various paths intersect with Returns on Investment;
and then we have figured out which is the optimal path to victory.
Anon @1:40
The saying "As Maine Goes, so goes the nation," dates from a comment made (author unknown) after the 1880 election and dervies from that fact that until recently Maine held its electiona (except for Presidential) in September. Thus Maine results for governor, senate, US house and the state legislature were known quite a bit before the main national election (when Maine also voted for president).
The September election was viewed as a portent of votes to come. The saying more or less died of ridicule after the the 1932 election, when Maine and Vermont were the only two states to vote for Alf Landon. In September the Republicans had won the governorship and the US House seats, and regained control of the legislature.
The ironing saying after then was, "As Maine goes, so goes Vermont."
Anon@7:53am: I wonder if the old saying "as Missouri goes, so goes the nation" will be wrong in 2008. Anyone know the last time MO and OH went for different candidates in a presidential election?
Haven't read the whole thread, so forgive me if this has already been answered.
The last times OH & MO differed on the presidential candidates was 1960 and 1956.
I list both years, because it's an interesting dichotomy. In both years, OH went R, and MO went D.
Which means that OH got the right pick in '56, but MO chose the winner in '60.
So, yes, there's precedent, albeit 50 years ago (+/- 2), for OH and MO to disagree, and each are equally likely to be correct. In other words, if they split, there's no reason to believe either has an edge in picking the winner.
.
Rhode Island: I liked that map: shows why colorado and montana favour Obama (skinny people like a a skinny guy). Also why the south is unlikely to turn blue - fat blacks are unlikely to turn out - the only 'audacious' hope is that they are 'outweighed' by fat whites staying home.
Nate,
There's an offensive Newsmax ad that intermittently appears at the bottom of the page.
It features a photo of Obama manipulated to look like a Mad magazine caricature of a black man circa 1973 - broadened nose, flattened cheek bones, etc.
It's really fucking racist.
I'm sure that's not the kind of content you want to see on your website.
Is there something you can do to block it, and register your displeasure with the advertiser?
.
If Newsmax wants to pay for my 538 addiction, they can put whatever nonsense they want at the bottom of the page!
I'd say the probability of an ad changing the vote of a 538 reader (in either direction) is ... what's the right term? ... "statistically irrelevant"?
One thing I would be interested in seeing:
We can view the data from this regression as giving us a weighted graph, where the vertices of the graph correspond to the states, and each edge is weighted according to the similarity.
For such graphs, there are algorithms based on spectral techniques to decompose the vertices into clusters of similar vertices. I'm curious what the results would look like when applied to this data set, especially if you remove the "latitude-longitude" variable from your regression and look purely on demographic/partisan index data.
JGabriel et al: thanks for your replies to my Ohio/Missouri question from 7:30 this morning. It will be interesting to see if they do, in fact, diverge this year.
Obsessed: I'd say the probability of an ad changing the vote of a 538 reader...
It's not whether the ad is effective or not, that's completely irrelevant.
It's that the ad is disturbingly and offensively racist. The Obama photo is clearly manipulated to resemble a cartoon caricature of a black man.
I suspect Nate doesn't want content like that appearing on his site - anymore than he'd want Stormfront advertising here - and called it to his attention.
.
Anon@3:07: JGabriel et al: thanks for your replies to my Ohio/Missouri question from 7:30 this morning. It will be interesting to see if they do, in fact, diverge this year.
Your welcome, Anon. I think it's interesting too.
I expect that if the election is close, we'll probably see a split between OH and MO. But if it turns into a large or landslide victory, then we'll probably see both selecting the same candidate.
.
Thank you, anon@12:29, for taking the time to explain the difference between North and South Carolina. Your comment regarding Northerners feeling more "welcome" in NC is interesting. It goes to what I brought up before: is there data on percentage of voters born outside the state v. born within?
(And please forgive me for the mild gripe earlier. One of the primary reasons I keep coming to this site--other than overall intelligence--is quality of design and EXCELLENT USE OF MAPS. Please don't think I don't appreciate it.)
Growing up in Connecticut, I can tell you that the state is very New England. You can look at the team loyalties (Red Sox, Patriots, Celtics) and figure that one out easily. The state is traditionally less liberal than the rest of New England (not including New Hampshire) though, and can be fairly said that the state is usually center-right in its politics. I do agree with what was pointed out though, that politically speaking, Connecticut is much closer to New Jersey than our other neighbors. Great map you have there! Well done.
Something I find interesting here is how few states form "pairs," where A is the state most like B, and B the state most like A. I think these are the only pairs:
AZ/NM
KY/TN
ME/VT
NJ/NY
NC/SC
ND/SD
OR/WA
No huge surprises there, I guess, but it makes for an interesting list!
Dear prematurely grey,
I probably should have admitted that I am a North Carolinian whose folks have lived there since leaving Europe, so my take on NC vs. SC is biased. But talk to black people or Northerners who know the two states, and you will probably hear similar things.
I would imagine NC has a higher % of people born out of state, but to reiterate my theory, NC's relative moderation and open-mindedness is a big reason it has boomed as a destination for new arrivals from the NE.
One of the ironies about Jesse Helms' legacy is that it tends to obscure the other strain in NC political history: the Democrats like Terry Sanford and Jim Hunt who helped make the research triangle park happen, and turned NC's public university system into one of the better ones (although it is fair to say that California is in a separate class). My point is, at the state level, North Carolina had leaders who saw the future and helped put the state in the excellent position it is in now (check out Charlotte and Raleigh in rankings of solid housing markets, education levels, job creation, etc). And the state's people (even some of us with - gasp - scotch-irish ancestry) deserve credit for making it a relatively welcoming destination.
I was wondering whether there was any state that could be regarded as archetype - something like the "centre of US politics" in a socio-demographic sense. Looking at Nate's table, it felt like MO, OH or MN could be it. Isabel's great "minimum-weight spanning tree" suggests that the answer should be OH. Nevertheless, it would be great if you, Nate, could run one extra turn of your similation to determine the state with the closest distance (or least-square distance, alternatively) to all other states. In any case, its interesting, but also no wonder (at second thought) that candidates for this "centre" are key swing and/or bellweather states, and ranking high on the tipping-point list.
@ jdk: I think, with the share of catholics, Nate is already including a pretty good proxy for your PI[i]GS in his list of variables. Asides from that, you are wrong when stating that the PI[i]GS (or a slightly extended ancestry mix) represent the non-WASP part of the poulation, since (a) the PI[i]GS do neither include Asian nor Native American population, which are both non-WASP as well [they also don't include AAs and Latino's, but Nate has several variables for those two], and (b) 19,1% of respondents did not state any ancestry at all, making them in fact the second largest ancestry group in the USA (after German).
I have (on limited subsamples, and on a county-by-county level) myself been investigating a bit on the relation between ancestry and Obama's vote share in the primaries. There is quite some indication about Italians having preferred Clinton (e.g. MA, CT, OH), as have been Hungarians (OH). For Portuguese, on the other hand, such a relation is less obvious. In fact, there are only two counties with significant Portuguese ancestry in the USA, both in RI, and they were the two in which Obama fared best during the State's primary. I tested the Irish on the example of Iowa (caution - caucus state!) and found no correlation, neither positive nor negative, between Irish ancestry share and the Obama vote share. English ancestry in Iowa countries tended to slightly reduce the Obama vote share, albeit also with low significance. If at all, it appears that both Irish and English ancestry resulted in a slight pro-Edwards tendency. French / Franco-Canadian ancestry appears to have favoured CLinton, while Arkadian / Cajun looks like having been pro-Obama. There is a slightly positive correlation of German and of Dutch ancestry shares with the Obama vote share, and a reasonable strong correlation of Scandinavian ancestry (all for Iowa). In practice, however, German, Dutch and Scandinavian ancestries are often so intervowen, and inter-marriage appears to have been so common, that there is hardly a point in differenting the three. If you look a "germanic" (i.e. German, Dutch, Swiss, Scandinavian combined)ancestry, you find quite a good correlation to the Obama vote share, though much less strong than for the "americans", which have R² values of 0.5 or higher in explaining the Clinton vote share.
As to the Polish: At the time I did my analyses, it was difficult to come out with meaningful results, because many areas east of the Missisippi with high Polish ancestry share are also having high German ancestry share, so it is difficult to single out any of the two. One might re-run the tests now by including states / counties with high German, but low Polish ancestry share (OR, WY, SD), which I have not done so far.
In any case, this all relates to the primary, and I have no idea to which extent the above patterns will still hold true for the GE. Nate's posting on the 'unity bounce' in Clinton strongholds indicates that at least some of the ancestry-related voting patterns may have to do more with a 'pro-Clinton' than with an 'anti-Obama' vote. I would have loved to see some respective updates from Nate on his regression results related to GE polling (nag, nag, nag), but he apparently has other priorities for the moment ..
PECOTA political analysis
THIS! IS THE ANALYSIS I WAS WAITING FOR!
/hope someone else here reads the dugout
**OnlineHost**
SteakGrowsOnDmitri is choking the hell out of PUMA_Voter.
That's for you, Ben.
Dear NC Anonymous,
I probably should have said that my people are in North Carolina.
@jsh1120: thanks! But the greatest signal to noise ratio on the net isn't saying much. I agree, though, that the comments here tend to be in general more insightful than most political blogs. I think that's especially true on posts like this one, which are difficult to interpret in a partisan way. The polling posts seem to be the ones that attract the McCain-supporter trolls. (Note that I'm not saying all McCain supporters are trolls, rather that most trolls on this site are McCain supporters.)
@Harrison: I understand wanting to look at things probabilistically. And I suspect that there might be a function f such that if the similarity score between two states is x, then the probability of them voting together (holding some suitable tings constant) is f(x). I just don't think it's linear, and since I don't know what's going on internally in the model I can't guess at what it is. (Even if I did have the output of whatever regression model is being used, I don't know if I could guess easily -- I'm a mathematician, not a statistician.)
@ajb: I think this is actually pretty typical of what happens if you just take a bunch of random points and look at which ones are closest to each other -- just because X is the closest point to Y doesn't mean Y is the closest point to X. Also, AZ/NM is particular interesting as these pairs go because not only are they close to each other, but they're very far from all the other states.
@Frank from Germany: For what it's worth, I deliberately emphasized the centrality of Ohio when designing the chart -- Ohio happens to have a lot of neighbors in that diagram and so I was forced to draw it that way. If I had taken a more principled approach to the whole thing I would have found some way to draw the tree that didn't depend on me doing things by eye, but then I would have had to actually do some research on those methods, since I'm not familiar with them.
(Also, does anyone reading this think that some sort of threaded comment system would be useful?)
Heh, until I noted the two different dates, I figured Nate had thrown up a "partisan" entry for everyone to flame upon, followed by this "data-focused" entry that's relatively free from bias. Then I noticed they were posted on two different days.
That doesn't mean it isn't a good idea, mind you....
As for the comment about the Newsmax ad, my impression (as a guy who's been a graphic designer for a decade for what it's worth) is that the photo hasn't been manipulated at all, it's just that the photo was taken at a lower angle relative to the man - you're looking up at him, slightly. If they wanted to be racist, they'd probably would have first darkened his skin like some of the flyers circulating around have done.
There's a similar photo here - foreshortening due to camera angle:
http://www.topnews.in/usa/files/Barack-Obama-Hillary-Clinton-New-Hampshire-Primaries.jpg
JGabriel, I visit this site embarrassingly often and I'm afraid I'm not sure what Newsmax ad you're referring to. The only ones I'm seeing right now are the one that says "Obama or McCain?" with a picture of each of them and another that says "Over for Obama?" with a similar picture of him. Neither of them looks racist in any way to me. What am I missing?
Dear Frank from Germany -
One problem with breaking up the US's white population into country of origin is that many of the early settlers took on the name "American" after a few decades. So, for example, it is very hard to tell what percentage of the population has some Scotch-Irish or Welsh ancestry, even in Appalachia, where they
settled most heavily, and I imagine it affects tracking down English, German, and French ancestry, too.
Intermarriage between all these white groups, and with other groups, is probably a reason for that "American" label taking hold as a substitute for one's
ethnic background. I should point out that this problem affects demographic research in much of the country, because these groups (the Scotch-Irish are probably the most obvious example) kept moving, and were among
the early settlers of the Deep South, Texas, and the West, too. There are some interesting maps about this on wikipedia on the Scotch-Irish and "American" ethnicity pages.
Costa: I visit this site embarrassingly often and I'm afraid I'm not sure what Newsmax ad you're referring to. The only ones I'm seeing right now are the one that says "Obama or McCain?" with a picture of each of them and another that says "Over for Obama?"
I believe it's the Newsmax ad that says "Over for Obama". It didn't come up this time, so I can't verify it, but I think that's the one.
Take a good look at the photo next time you see it. Obama's nose is broadened and flattened, and a shine appears to be added to his nose and forehead. Also, his face is broadened; his unmanipulated face is narrower. There's also a kind of waviness or texturization that's noticable around, or just below, the temples.
If I'm right, it's the equivalent of Time (or was it Newsweek?) darkening OJ Simpson's face for it's cover a few years ago to make him look scarier - only Newsmax's manipulation of the Obama photo is more elaborate.
Anon@4:26pm seems to disagree with that analysis, but the photo he submits as an example still shows a narrower face and lighter skin (though that could be due to lighting differences in the photos). Frankly, I didn't find it a very convincing counter-example.
.
.
JGabriel:
WTF are you talking about the Newsmax ad "Over for Obama" could simply be a photo taken with a short focal length lens that tend to shown ones characteristics fuller rather than flat (which telephoto lenses do).
Your post is bs imo. And btw I think Obama looks fine for in the photo
Dear Anon above: The issue with the "Americans" is not only one of early settlement. Otherwise, you should find a lot of "Americans" in New England, which you don't. Instead, most people there are pretty well able/ willing to declare themselves in the Census as being of English or French ancestry. You also find a surprisingly high number of people with "Pennsylvania German" ancestry in the US Census.
In fact, I think there is hardly a country in the world which is as obsessed with ancestry research and genealogy as is the USA. And, from what I have learned (corresponding with some Americans researching back on their German roots, or looking for hints on the origin of their family name), there is quite some documentation available for such research, including old US census lists, shipping lists, immigration records, etc.
So (in my ínterpretation), the fact that somebody in the USA is either unwilling or unable to state his/her ancestry, and/or just calls it American is more due to socio-political factors. It may be a political statement, which implies neglecting ancestors' own immigration history and is thus in tendency anti-immigrant. It may reflect missing family records and memorables (photographs, letters etc.) and lack of respective interest, which is not neccessarily, but may often be co-inciding with low education and weak social status.
Or, it relates to some sort of shame on disclosing the ancestry, if it is actually known. Check, e.g. "Melungeon" on Wikipedia. From what I have read, it as well appears that part of the early British settlement to the South and especially to Appalchia was not voluntary, but consisting of deported prisoners and orphans that were forced into mining work (similarly to the British practice when settling Australia in the 19th century). Such a family history
is quite likely to not being transferred over the generations, and even if it were known to you, you would not neccessarily disclose it in the Census.
In any case, it can be statistically demonstrated that, during the primaries, self-declared "Americans" voted differently than people with self-declared Scotch-Irish ancestry. As to why this has been the case is surely a fascinating topic for socio-anthrological research.
Frank from Germany said...
I was wondering whether there was any state that could be regarded as archetype - something like the "centre of US politics" in a socio-demographic sense.
In previous studies, depending which demographics were considered, usually Illinois, Wisconsin, and Missouri turn up as being the most representative of the nation as a whole. Same ethnic, religious, economic breakdowns. Same education levels. Same ratios of urban to suburban to rural residents. Same party identification breakdowns. Etc.
Personally, I find the diversity of Illinois to be more compelling as a microcosm of the nation as a whole. However, considering that Obama is from Illinois, that probably makes it less effective as a model for this election cycle. For this election year, I think Missouri would be a good archtype.
Pretty chart, now if only I could decipher it's meaning. What do the colors refer to?
Frank from Germany wrote (in response to my Anon comments above):
"So (in my ínterpretation), the fact that somebody in the USA is either unwilling or unable to state his/her ancestry, and/or just calls it American is more due to socio-political factors. It may be a political statement, which implies neglecting ancestors' own immigration history and is thus in tendency anti-immigrant. It may reflect missing family records and memorables (photographs, letters etc.) and lack of respective interest, which is not neccessarily, but may often be co-inciding with low education and weak social status.
Or, it relates to some sort of shame on disclosing the ancestry, if it is actually known. Check, e.g. 'Melungeon' on Wikipedia."
First, big-time props for someone in another country knowing who the Melungeons are - that's a very interesting piece of Southern/Appalachian history. I didn't know anyone outside of NC or TN know about them. That said, I don't think biracial/triracial groups like that make up much of the large "American" ancestry grouping, in part because census takers probably would have marked anyone they considered black or Indian as members of those groups under the one-drop rule (if they could find them - these groups often were kind of excluded from normal life), and many mixed race people who could have "passed" for white probably would have played up whatever white national origin they could claim.
I do think you're probably right about the political statement, and economic status, parts of it. A lot of early settlers to the South/Appalachia were poor, some of them had been in prison back in Europe, and some of them had religious reasons for leaving as well (in the Scotch-Irish case, in addition to poverty, the British banned Presbyterian baptisms in Ulster around 1700 or 1710). And then, when these people got to America, guess who controlled the Eastern coastal establishments? British Governors and colonists - meet the new boss, same as the old boss, if you were coming from Ulster. In NC, for example, there was a harsh divide between the more British coastal aristocracy and the more Scotch-Irish and German western areas (well, the central and western areas based on modern notions of NC geography). Basically, to use Obama's unfortunate term about (heavily Scotch-Irish and German central) PA, these white settlers were probably bitter, and wanted to be part of something new, in a way that more economically-stable New Englanders may not have felt.
I would also add a racial angle - these Appalachian settlers who were constantly fighting with Indian tribes at the country's western borders probably thought about ethnicity primarily in terms of "us" vs. "Indians." The Indian Wars (and slavery) may have worked to merge these somewhat diverse groups of white settlers into the "white American" ancestry grouping (along with intermarriage between the various groups).
Purchase Acyclovir.Buy cheap acyclovir.Cheap acyclovir.Acyclovir 400mg
Buy Atenolol.Order Atenolol.Generic Atenolol.Cheap Atenolol
Order Augmentin. Generic Augmentin. Order augmentin online.Buy augmentin
Order Azithromycin.Cheap azithromycin.Order azithromycin online.purchase azithromycin.buy azithromycin.
Buy Avandia.Generic avandia.Purchase avandia online.Avandia 4mg
Generic Biaxin.Order Biaxin.Cheap biaxin.Order biaxin online
Order Cipro online.Buying Cipro.Purchase Cipro.Discout Cipro
Buy claritin.Generic Claritin.Buy claritin d.Order claritin
Buy Cozaar.Cozaar generic.Order cozaar.Cozaar 50mg
Purchase Diflucan.Buy cheap diflucan.Diflucan 150mg
It might be interesting to try to use some version of this to calculate a covariance matrix for poll movement among the states. Then, instead of having separate national and state-specific random variables in your Monte Carlo simulation, you could simply have multivariate normal random variables with the calculated covariances. This would seem to eliminate the concern about double counting, though it would be susceptible to errors in your covariance model.
NC also differs from SC in that there are some serious "liberal enclaves" here: Asheville, Boone, much of Durham, and particularly Chapel Hill/Carrboro. While the traditional associations of "The South" do fit for parts of North Carolina, the state is increasingly heterogenous. Outside of the ultra-lib zones and the conservative corners, you've got major cities that feel like... well, cities. Charlotte feels "Southernish" in the way Atlanta does, I guess. But the "Research Triangle" (Raleigh/Durham and surrounds) feels like Dulles Corridor or even maybe Route 128 (warmer, and with a different dialect, and crappy public transit).
The tech sector is one major cultural difference b/t NC and SC, I think. In NC, you have the birthplace of the Usenets, home of Red Hat, and a veritable explosion of tube-related work. In the center of the state, everything is telecom, biotech, IT work - nobody is manufacturing or farming or nuthin. Lots of evil scientists, too.
NC has good public education, not just at the university level, but starting from pre-K (4-term Governor Jim Hunt put major bucks into the "Smart Start" program to ensure that all NC kids start kindergarten with adequate health and educational support).
And even though we are best known for the Helms, NC also has some way lefty pols. Mel Watts and David Price are consistently left of the party leadership in the House. Oh, and just for weirdness, NC gave the Dems their current delegate / superdelegate primary voting structure (via the "Hunt Commission," on which Price served as staff director). So there is also a strong "blue" tradition in the state, in spite of its history of voting red in presidential and (most) senate contests.
Okay, sorry for the long post. To recap, the principle NC/SC differences: NC is more urban and techie, puts a lot of emphasis on schooling, and has traditional, committed pockets (both geographic and legislative) of hardcore liberal dems.
@Frank from Germany.
"@ jdk: I think, with the share of catholics, Nate is already including a pretty good proxy for your PI[i]GS in his list of variables." Not really. But I happen to know more about this stuff than almost anyone
"Asides from that, you are wrong when stating that the PI[i]GS (or a slightly extended ancestry mix) represent the non-WASP part of the poulation, since (a) the PI[i]GS do neither include Asian nor Native American population, which are both non-WASP."
First, my model of PIiGs includes certain Asian ethnic communities on the West Coast.
In a post-racial US, it may be that a BASP BlackAngloSpeakingProt. passes as a WASP after you control for age and income.
As to individual ancestry testing, it is more complicated than that which is why a combined value variable makes sense, but how I do it is with some non-continuous functions. E.g. Polish below 3.1% code as 0, Portuguese below 4.0% code as 0, Italian below 4.5% code as 0. Etc. And/or transform the data to Z scores based on national dispersion. Etc.
Using a single an ethnic index variable (which doesn't even use the non-continuous functions described), I can get an R2 almost of .70 to predict Kerry percentage of vote, excluding some difficult outliers like DC, UT, WY in the 2004 election.
To more fully explain would be to aid McCain.
Since you have similarity/dissimilarity scores, why not run it though a multidimensional scaling (MDS), so that you can get a map? The first two components will give a 2D map (and almost certainly cover most of the variation captured by your variables).
In this instance the wikipedia page on MDS is not a lot of help, but some other pages are somewhat useful, for example:
http://www.analytictech.com/borgatti/mds.htm
http://www.statsoft.com/textbook/stmulsca.html
http://forrest.psych.unc.edu/teaching/p208a/mds/mds.html
The calculations required for MDS are not onerous (just a couple of lines in Matlab), but in any case, many statistical packages can perform MDS automatically, just using as input a matrix of either distances or similarities between pairs of objects (which you have).
If you don't have it already, the free statistical package /R/ (http://cran.r-project.org/) has MDS built-in (via "cmdscale") and also has an add-on package available that does stress minimization by means of majorization ("smacof").
This page (http://www.personality-project.org/r/mds.html) shows how to do MDS in R.
I've created an ethno-political map here using a similar method:
http://kromkowski.blogspot.com/
This was quickly calculated. But
"ethnicity" is obviously a pretty salient factor.
538 is primarily about prediction. But such analysis also suggests strategic decisions. I don't mean Marc Penn voodoo micro-trend bs or slicing and dicing. Rather, the importance of the "ethnic factor" suggest sensitivity to a pluralistic vision of America that is beyond narrow black/white "racialism" or simplistic black/white/hispanic/asian models.
Beyond race and toward multi-cultural competence
Senator Obama is NOT a descendant of slaves, but an immigrant father’s son. As a catholic Polish Italian American, I don’t connect through his “white” Kansas grandmother who admitted fearing young blacks. I connect because his immigrant father makes him an “ethnic”.
Transcending “race” is not ignoring racial discrimination. But “overcoming” requires abandoning the dualistic black-white "race" language and replacing a pluralistic language of "ethnicity". The late Msgr. Geno Baroni marched with Rev. King and was an organizing voice of the so-called “white ethnic” Neighborhood movement of the 1970s. In 1968, he told American Bishops that little attention was given to the anguish of socially and politically alienated ethnic Americans – descendants of largely but not exclusively catholic immigrants, who were burdened by taxation, dissatisfied with government, fearful of job security and concerned with prohibitive college costs. Baroni claimed that “ethnicity” for these Americans was what gave their children both “roots and wings”. It was a dimension of social justice that couldn’t be overlooked. It is no little coincidence that Obama's first organizing work was funded by the Campaign for Human Development which was in good part Baroni's brainchild.
Race" does not give "roots and wings", it is a political concept used to exclude others.
“Ethnicity” is a familial, psychological and cultural construct which begins with a place – the home – expanding physically, geographically and emotionally to include others.
The confusion has been that “black” or “African American” is a statement of both race and ethnicity. Paired with “white” it’s racial; paired with “Polish American”, it’s ethnicity.
Today, ethnicity is not the “food and festivals” ethnicity of the 1970s. The “new ethnicity” includes ancestry but is broader. It includes cheering for "your team", neighborhood parties, Aunt Anne's three-bean salad, and all family, neighborhood, and community traditions which give our children “roots and wings”.
This more modern attempt at ethno-political mapping shows that we're all "ethnic", even those who choose "American" or choose not to respond to the Ancestry question.
Senator Obama is a product of a mixed marriage, but everyone is a product of a "mixed" marriage. The Notre Dame fan marries the Southern Cal fan. The Irish American marries the German American. Your mother married your father. Ethnic identity is created by addition and iteration. It provokes the questions: “Who am I and Who are We?” as a way to learn how to love ourselves, our neighbors, our world.
I know that McCain has no vision to understand this nuance.
I am confident that Obama having worked in Chicago and being the son of immigrant must understand it. Part of winning will require him to transmit his sensitivity to ethnicity as the paradigm to replace mere racialism.
In light of Biden, one might care to notice in the paper here:
http://kromkowski.blogspot.com/
how in my nearest cultural neighbor analysis, DE is the link to VA and to MI in a minimum tree spanning network.
I decided to make a directed graph of the data, with each state pointing to its top match. The result is a set of disjoint trees, all of whom end in a paired set. Each tree consists of a subgroup of states, and those subgroups are kind of interesting:
1. ME, VT
2. AZ, NM
3. GA, SC, NC
4. DC, MA, CT, NJ, RI, NY
5. AR, MS, OK, AL, TN, WV, KY
6. UT, ID, WY, MT, KS, NE, ND, SD
7. TX, MD, VA, DE, FL, LA, MO, PA, IN, OH, MI
8. CA, IA. NH, IL, WI, MN, AK, HI, NV, CO, WA, OR
The two biggest groups are the most surprising: TX and DE? IL and AK? Of course, the groupings don't necessarily mean anything (other than in a six-steps-to-Kevin-Bacon sort of way), but it was fun.
(I can post the graph itself if anyone's interested.)
Can anyone see the image or does anyone have a copy of it?
徵信社, 感情挽回, 挽回感情, 徵信, 徵信社, 徵信, 捉姦, 徵信公司, 通姦, 通姦罪, 抓姦, 抓猴, 捉猴, 捉姦, 監聽, 調查跟蹤, 反跟蹤, 外遇問題, 徵信, 捉姦, 女人徵信, 外遇問題, 女子徵信, 外遇, 徵信公司, 徵信網, 徵信, 徵信社, 外遇蒐證, 抓姦, 抓猴, 捉猴, 調查跟蹤, 反跟蹤, 感情挽回, 挽回感情, 外遇沖開, 徵信, 徵信, 徵信社, 抓姦, 徵信, 徵信社, 外遇蒐證, 外遇, 通姦, 通姦罪, 贍養費, 徵信, 徵信社, 徵信社, 抓姦, 徵信社, 徵信社, 徵信, 徵信社, 徵信, 徵信, 徵信公司, 徵信社, 徵信, 徵信公司, 徵信社, 徵信社, 徵信社, 徵信社, 徵信社, 徵信, 徵信公司, 徵信社, 徵信, 徵信, 徵信公司, 徵信, 徵信社, 女人徵信
徵信, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵信, 徵信網, 徵信社, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵信,徵信網, 徵信, 徵信社, 外遇, 徵信, 徵信, 徵信社, 抓姦, 徵信, 徵信社, 外遇, 徵信, 徵信社, 徵信社, 徵信社, 抓姦, 徵信社, 徵信社, 徵信, 徵信社, 徵信, 徵, 徵信公司, 徵, 徵, 徵, 徵, 徵, 徵, 徵信, 徵信公司, 徵信社, 徵, 徵, 徵, 徵信, 徵信公司, 徵信社, 徵信, 徵, 徵, 徵, 徵, 徵信, 徵信社, 徵信公司, 徵, 徵, 徵, 徵信, 徵信社, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵, 徵
情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,情趣,美國aneros,rudeboy,英國rudeboy,英國Rocksoff,德國Fun Factory,Fun Factory,英國甜筒造型按摩座,甜筒造型按摩座,英國Rock Chic ,瑞典 Lelo ,英國Emotional Bliss,英國 E.B,荷蘭 Natural Contours,荷蘭 N C,美國 OhMiBod,美國 OMB,Naughti Nano ,音樂按摩棒,ipod按摩棒,美國 The Screaming O,美國TSO,美國TOPCO,美國Doc Johnson,美國CA Exotic,美國CEN,美國Nasstoy,美國Tonguejoy,英國Je Joue,美國Pipe Dream,美國California Exotic,美國NassToys,美國Vibropod,美國Penthouse,仿真按摩棒,矽膠按摩棒,猛男倒模,真人倒模,仿真倒模,PJUR,Zestra,適趣液,穿戴套具,日本NPG,雙頭龍,FANCARNAL,日本NIPPORI,日本GEL,日本Aqua Style,美國WET,費洛蒙,費洛蒙香水,仿真名器,av女優,打炮,做愛,性愛,口交,吹喇叭,肛交,魔女訓練大師,無線跳蛋,有線跳蛋,震動棒,震動保險套,震動套,TOY-情趣用品,情趣用品網,情趣購物網,成人用品網,情趣用品討論,成人購物網,鎖精套,鎖精環,持久環,持久套,拉珠,逼真按摩棒,名器,超名器,逼真老二,電動自慰,自慰,打手槍,仿真女郎,SM道具,SM,性感內褲,仿真按摩棒,pornograph,hunter系列,h動畫,成人動畫,成人卡通,情色動畫,情色卡通,色情動畫,色情卡通,無修正,禁斷,人妻,極悪調教,姦淫,近親相姦,顏射,盜攝,偷拍,本土自拍,素人自拍,公園露出,街道露出,野外露出,誘姦,迷姦,輪姦,凌辱,痴漢,痴女,素人娘,中出,巨乳,調教,潮吹,av,a片,成人影片,成人影音,線上影片,成人光碟,成人無碼,成人dvd,情色影音,情色影片,情色dvd,情色光碟,航空版,薄碼,色情dvd,色情影音,色情光碟,線上A片,免費A片,A片下載,成人電影,色情電影,TOKYO HOT,SKY ANGEL,一本道,SOD,S1,ALICE JAPAN,皇冠系列,老虎系列,東京熱,亞熱,武士系列,新潮館,情趣用品,情趣,情趣商品,情趣網站,跳蛋,按摩棒,充氣娃娃,自慰套,G點,性感內衣,情趣內衣,角色扮演,生日禮物,生日精品,自慰,打手槍,潮吹,高潮,後庭,情色論譠,影片下載,遊戲下載,手機鈴聲,音樂下載,開獎號碼,統一發票號碼,夜市,統一發票對獎,保險套,做愛,減肥,美容,瘦身,當舖,軟體下載,汽車,機車,手機,來電答鈴,週年慶,美食,徵信社,網頁設計,網站設計,室內設計,靈異照片,同志,聊天室,運動彩券,大樂透,威力彩,搬家公司,除蟲,偷拍,自拍,無名破解,av女優,小說,民宿,大樂透開獎號碼,大樂透中獎號碼,威力彩開獎號碼,討論區,痴漢,懷孕,美女交友,交友,日本av,日本,機票,香水,股市,股市行情, 股市分析,租房子,成人影片,免費影片,醫學美容,免費算命,算命,姓名配對,姓名學,姓名學免費,遊戲,好玩遊戲,好玩遊戲區,線上遊戲,新遊戲,漫畫,線上漫畫,動畫,成人圖片,桌布,桌布下載,電視節目表,線上電視,線上a片,線上掃毒,線上翻譯,購物車,身分證製造機,身分證產生器,手機,二手車,中古車,法拍屋,歌詞,音樂,音樂網,火車,房屋,情趣用品,情趣,情趣商品,情趣網站,跳蛋,按摩棒,充氣娃娃,自慰套, G點,性感內衣,情趣內衣,角色扮演,生日禮物,精品,禮品,自慰,打手槍,潮吹,高潮,後庭,情色論譠,影片下載,遊戲下載,手機鈴聲,音樂下載,開獎號碼,統一發票,夜市,保險套,做愛,減肥,美容,瘦身,當舖,軟體下載,汽車,機車,手機,來電答鈴,週年慶,美食,徵信社,網頁設計,網站設計,室內設計,靈異照片,同志,聊天室,運動彩券,,大樂透,威力彩,搬家公司,除蟲,偷拍,自拍,無名破解, av女優,小說,民宿,大樂透開獎號碼,大樂透中獎號碼,威力彩開獎號碼,討論區,痴漢,懷孕,美女交友,交友,日本av ,日本,機票,香水,股市,股市行情,股市分析,租房子,成人影片,免費影片,醫學美容,免費算命,算命,姓名配對,姓名學,姓名學免費,遊戲,好玩遊戲,好玩遊戲區,線上遊戲,新遊戲,漫畫,線上漫畫,動畫,成人圖片,桌布,桌布下載,電視節目表,線上電視,線上a片,線上a片,線上翻譯,購物車,身分證製造機,身分證產生器,手機,二手車,中古車,法拍屋,歌詞,音樂,音樂網,借錢,房屋,街頭籃球,找工作,旅行社,六合彩,水噹噹,台中隆鼻,果凍隆乳,改運整型,自體脂肪移植,新娘造型,婚禮顧問,下川島,常平,常平,珠海,澳門機票,香港機票,貸款,貸款,信用貸款,宜蘭民宿,花蓮民宿,未婚聯誼,網路購物,婚友,婚友社,未婚聯誼,交友,婚友,婚友社,單身聯誼,未婚聯誼,未婚聯誼, 婚友社,婚友,婚友社,單身聯誼,婚友,未婚聯誼,婚友社,未婚聯誼,單身聯誼,單身聯誼,白蟻,白蟻,除蟲,老鼠,減肥,減肥,在家工作,在家工作,水噹噹,台中隆鼻,果凍隆乳,改運整型,自體脂肪移植,新娘造型,婚禮顧問,下川島,常平,常平,珠海,澳門機票,香港機票,貸款,貸款,信用貸款,宜蘭民宿,花蓮民宿,未婚聯誼,網路購物,婚友,婚友社,未婚聯誼,交友,婚友,婚友社,單身聯誼,未婚聯誼,未婚聯誼, 婚友社,婚友,婚友社,單身聯誼,婚友,未婚聯誼,婚友社,未婚聯誼,單身聯誼,單身聯誼,白蟻,白蟻,除蟲,老鼠,減肥,減肥,在家工作,在家工作,婚友,單身聯誼,未婚聯誼,婚友,交友,交友,婚友社,婚友社,婚友社,大陸新娘,大陸新娘,越南新娘,越南新娘,外籍新娘,外籍新娘,台中坐月子中心,搬家公司,搬家公司,中和搬家,台北搬家,板橋搬家,新店搬家,線上客服,網頁設計,線上客服,網頁設計,植牙,關鍵字,關鍵字,seo,seo,網路排名,自然排序,網路排名軟體,交友,越南新娘,婚友社,外籍新娘,大陸新娘,越南新娘,交友,外籍新娘,視訊聊天,大陸新娘,婚友社,婚友,越南新娘,大陸新娘,越南新娘,視訊交友,外籍新娘,網路排名,網路排名軟體,網站排名優化大師,關鍵字排名大師,網站排名seo大師,關鍵字行銷專家,關鍵字,seo,關鍵字行銷,網頁排序,網頁排名,關鍵字大師,seo大,自然排名,網站排序,網路行銷創業,汽車借款,汽車借錢,汽車貸款,汽車貸款,拉皮,抽脂,近視雷射,隆乳,隆鼻,變性,雙眼皮,眼袋,牙齒,下巴,植牙,人工植牙,植髮,雷射美容,膠原蛋白,皮膚科,醫學美容,玻尿酸,肉毒桿菌,微晶瓷,電波拉皮,脈衝光,關鍵字,關鍵字,seo,seo,網路排名,自然排序,網路排名軟體,英語演講,托福,Toastmaster,汽車借款,汽車借款,汽車借款,汽車貸款,汽車貸款,借錢,借貸,當舖,借款,借貸,借錢,週轉,
Proposal For 2012 Primaries
http://www.huffingtonpost.com/will-bower/proposal-for-2012-primari_b_142976.html
From December 2007 to March 2008, I wrote various drafts of a proposal on how our political parties -- starting in 2012 -- might adopt primary election procedures that would better serve our country in selecting presidential candidates. I originally drafted a hypothetical calendar for 2008, based on general election results from 2004. Now that we have the results for 2008, I can now propose a calendar specific to 2012.
The system by which our parties choose their presidential candidates has proven itself to be, at best, highly questionable -- at worst, severely flawed.
The primary calendar we need most is one that is built on an orderly and rational plan -- one that is based on mathematics and on recent historical outcomes -- and not on an arbitrary, publicity-driven, system of one-upsmanship. The change I propose would provide for a more effective, equitable process than the one we have now.
The following factors are the key ones to consider:
Margin of Victory
- The state primaries would be placed in order according to the leading candidates' margins of victory in the preceding general election -- with the states registering the closest margins of victory going first.
For example, John McCain won Missouri by 0.1% and Barack Obama won North Carolina by 0.4%; conversely, McCain won Wyoming by 33%, and Obama won Hawaii by 45%. Therefore, the primary calendar I propose would commence with primaries being held in states such as Missouri and North Carolina -- and would close with such states as Wyoming and Hawaii.
- The purpose of ordering the states according to the margin of victory is to help the parties determine which candidates can appeal to those states that have found themselves most recently on the Electoral Divide. A narrow margin in the general election is reflective of an evenly divided electorate. In this scenario, a candidate who appeals to, say, Florida and Montana is more likely to appeal to a greater number of Americans on the whole.
Iowa, New Hampshire, and Fairness
- Iowa and New Hampshire might object to this new system, given their longstanding tradition of being the first states to cast their ballots. However, so long as Iowa and New Hampshire retain their record of being fairly bipartisan states, they'll maintain their position towards the front of the primary schedule.
- Just because a state should have its primary later in the season does not mean that that state will prove invaluable to the process. Indiana and North Carolina weren't held until May 6th, but those two states might have very well decided the fate of the 2008 Democratic nomination.
- This new system allows other states to play a greater role in how the parties select their candidates. For example, Missouri and North Carolina would be two of the states to get the limelight in 2012. Likewise, based on the results to come in November of 2012, a still-different slate of states could have a more significant role come 2016. A rotating system will be healthier and fairer.
Groupings of Five, and Timing & Spacing
- By placing states into groupings of five, no one state will be overly emphasized on any given date.
- Candidates will still need to address the concerns of individual states, whilst having to maintain an overall national platform. For example, a candidate will be less able to campaign against NAFTA in Ohio whilst campaigning for it in Florida.
- Given that each state has its own system for electing its delegates, these groupings of five states will act as an overall balancer. Ideally, caucuses will be done away with altogether by 2012. However -- should that not happen -- states with caucuses, states with open primaries, and states with closed primaries can all coexist within a grouping, therefore no one system will hold too much influence on any given date.
- Racial and geographic diversity in this process has been a great concern for many. The narrowest margins of victory in 2008 were in a wide variety of regions -- the Midwest, the Great Lakes, the Mid-Atlantic, the South, and the West.
- All parties would have an interest in addressing these narrow-margined states early on. The incumbent will want to win over those states that were most in doubt of him in the previous election, and opposing parties will want to put forth candidates who have the best chance of winning over those very same states.
- Primaries will be held biweekly, giving candidates and the media enough time to process and respond to the outcomes of each wave of primaries.
- Washington DC will be placed in the same grouping as whichever state -- Virginia or Maryland -- is closer to its own margin of victory.
- American Samoa, Guam, Puerto Rico, and Americans Abroad -- not having Electoral votes of their own -- will determine their own primary dates, so long as they occur between the first grouping and the last grouping.
Under these guidelines, the proposed calendar for the 2012 primary season is:
January 2012
Tue, 1/10
Missouri
North Carolina
Indiana
Florida
Montana
Tue, 1/24
Ohio
Georgia
Virginia
Colorado
South Dakota
Tue, 2/7
North Dakota
Arizona
South Carolina
Iowa
New Hampshire
Tue, 2/21
Minnesota
Pennsylvania
Texas
Nevada
West Virginia
Tue, 2/26
Mississippi
Wisconsin
New Jersey
New Mexico
Tennessee
Tue, 3/6
Kansas
Nebraska
Oregon
Kentucky
Michigan
Tue, 3/20
Washington
Maine
Louisiana
Arkansas
Alabama
Tue, 4/3
Connecticut
California
Illinois
Delaware
Maryland
Washington DC
Tue, 4/17
Alaska
Idaho
New York
Massachusetts
Rhode Island
Tue, 5/1
Utah
Oklahoma
Wyoming
Vermont
Hawaii
^^ nice blog!! ^@^
徵信, 徵信, 徵信, 徵信社, 徵信社, 徵信社, 感情挽回, 婚姻挽回, 挽回婚姻, 挽回感情, 徵信, 徵信社, 徵信, 徵信, 捉姦, 徵信公司, 通姦, 通姦罪, 抓姦, 抓猴, 捉猴, 捉姦, 監聽, 調查跟蹤, 反跟蹤, 外遇問題, 徵信, 捉姦, 女人徵信, 女子徵信, 外遇問題, 女子徵信, 徵信社, 外遇, 徵信公司, 徵信網, 外遇蒐證, 抓姦, 抓猴, 捉猴, 調查跟蹤, 反跟蹤, 感情挽回, 挽回感情, 婚姻挽回, 挽回婚姻, 外遇沖開, 抓姦, 女子徵信, 外遇蒐證, 外遇, 通姦, 通姦罪, 贍養費, 徵信, 徵信社, 抓姦, 徵信社, 徵信, 徵信公司, 徵信社, 徵信, 徵信公司, 徵信社, 徵信公司, 女人徵信, 外遇
徵信, 徵信網, 徵信社, 徵信網, 外遇, 徵信, 徵信社, 抓姦, 徵信, 女人徵信, 徵信社, 女人徵信社, 外遇, 抓姦, 徵信公司, 徵信社, 徵信社, 徵信社, 徵信社, 徵信社, 徵信社, 女人徵信社, 徵信社, 徵信, 徵信社, 徵信, 女子徵信社, 女子徵信社, 女子徵信社, 女子徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社,
^^ nice blog!! thanks a lot! ^^
徵信, 徵信社,徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 外遇, 抓姦, 離婚, 外遇,離婚,
徵信, 外遇, 離婚, 徵信社, 徵信, 外遇, 抓姦, 徵信社, 徵信, 徵信社, 徵信, 外遇, 徵信社, 徵信, 外遇, 抓姦, 徵信社, 征信, 征信, 徵信, 徵信社, 徵信, 徵信社, 征信, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社, 徵信社, 徵信社, 徵信, 外遇, 抓姦, 徵信, 徵信社, 徵信, 徵信社,
Post a Comment