Where the Whiter Folk Are
Today I combined some Census data with 2008 election results (thanks Cosma). Though Barack Obama won the vote last fall, he lost the Non-Hispanic white vote. It stands to reason then that the whiter and less Hispanic a county is, the more likely it would be to tilt McCain. I was curious as to geographic variation within this general rule-of-thumb. So I plotted the % who voted for Obama in a county vs. the % who were Non-Hispanic whites (according to the 2000 Census*). I then generated a line of best fit via loess, and used the deviations from the trend to generate a map shaded proportionately. In other words, the bluer a county is the more it voted for Obama above expectation based on the overall relationship of the % white Non-Hispanic within a county and vote for Obama (the converse for red naturally). Again, click the map for the larger image.
I also decided to constrain the data set to those counties which were at minimum 80% Non-Hispanic white. Mostly because the “Black Belt” counties are showing up on the above map.
Finally, a shaded map of the results.
1) Don’t mess with Texas Obama. Despite Obama winning the Hispanic counties in the South, those counties are always underperforming relative to other majority-minority districts. Since some of those counties are 90% Hispanic it probably isn’t just Non-Hispanic white swing in the other direction.
2) Though Obama lost much of the rural North and East, he overperformed when you use the whole nation as a reference point, in particular rural areas in the West and South.
3) Pennsylvania kind of does look like Pittsburgh + Philadelphia, with Alabama in the middle.
4) Obama did well in Greater New England. Less well in the Butternut Region of the southern Midwest settled from the South.
Here are a list of the 200 counties furthest from the trendlines. The first 100 are the most pro-Obama above expectation when the predictor is Non-Hispanic white %. The last 100 the most pro-McCain.
| Deviation From Trend Line | % For Obama | County & State |
| 0.4 | 0.78 | Marin, California |
| 0.39 | 0.77 | Multnomah, Imbler |
| 0.38 | 0.78 | Santa Cruz, California |
| 0.38 | 0.77 | San Miguel, Colorado |
| 0.36 | 0.74 | Sonoma, California |
| 0.36 | 0.75 | Dukes, Massachusetts |
| 0.35 | 0.75 | Berkshire, Massachusetts |
| 0.35 | 0.74 | Pitkin, Colorado |
| 0.34 | 0.73 | Dane, Wisconsin |
| 0.34 | 0.72 | Orange, North Carolina |
| 0.34 | 0.72 | Boulder, Colorado |
| 0.33 | 0.73 | Windham, Vermont |
| 0.33 | 0.73 | Franklin, Massachusetts |
| 0.33 | 0.72 | Hampshire, Massachusetts |
| 0.32 | 0.7 | Washtenaw, Michigan |
| 0.32 | 0.7 | Mendocino, California |
| 0.32 | 0.7 | King, Washington |
| 0.31 | 0.71 | Lamoille, Vermont |
| 0.31 | 0.84 | San Francisco, California |
| 0.31 | 0.7 | New Castle, Delaware |
| 0.31 | 0.7 | Johnson, Iowa |
| 0.31 | 0.69 | Tompkins, New York |
| 0.3 | 0.7 | Washington, Vermont |
| 0.3 | 0.7 | San Juan, Washington |
| 0.3 | 0.69 | Imbler, Wisconsin |
| 0.3 | 0.77 | Suffolk, Massachusetts |
| 0.3 | 0.83 | Philadelphia, Pennsylvania |
| 0.3 | 0.72 | Arlington, Virginia |
| 0.3 | 0.69 | Windsor, Vermont |
| 0.29 | 0.69 | Addison, Vermont |
| 0.29 | 0.69 | Silver Bow, Montana |
| 0.29 | 0.68 | Nantucket, Massachusetts |
| 0.29 | 0.72 | Montgomery, Maryland |
| 0.29 | 0.69 | Cuyahoga, Ohio |
| 0.28 | 0.68 | Chittenden, Vermont |
| 0.28 | 0.66 | Ingham, Michigan |
| 0.28 | 0.66 | Ramsey, Minnesota |
| 0.28 | 0.75 | Denver, Colorado |
| 0.28 | 0.67 | Iowa, Wisconsin |
| 0.28 | 0.67 | Camden, New Jersey |
| 0.27 | 0.67 | Deer Lodge, Montana |
| 0.27 | 0.66 | Summit, Colorado |
| 0.27 | 0.66 | Blaine, Idaho |
| 0.27 | 0.65 | Lucas, Ohio |
| 0.27 | 0.65 | Genesee, Michigan |
| 0.27 | 0.65 | Hartford, Connecticut |
| 0.27 | 0.66 | Monroe, Indiana |
| 0.27 | 0.66 | Jefferson, Washington |
| 0.27 | 0.66 | Bennington, Vermont |
| 0.26 | 0.66 | Athens, Ohio |
| 0.26 | 0.76 | Durham, North Carolina |
| 0.26 | 0.66 | Douglas, Wisconsin |
| 0.26 | 0.68 | Milwaukee, Wisconsin |
| 0.26 | 0.67 | Mercer, New Jersey |
| 0.26 | 0.65 | Benton, Oregon |
| 0.26 | 0.76 | Cook, Illinois |
| 0.26 | 0.64 | Hennepin, Minnesota |
| 0.26 | 0.64 | Muskegon, Michigan |
| 0.26 | 0.65 | Napa, California |
| 0.26 | 0.64 | Middlesex, Massachusetts |
| 0.26 | 0.64 | Douglas, Kansas |
| 0.26 | 0.77 | Santa Fe, New Mexico |
| 0.26 | 0.74 | Wayne, Michigan |
| 0.26 | 0.65 | Orange, Vermont |
| 0.26 | 0.74 | San Mateo, California |
| 0.25 | 0.65 | St. Louis, Minnesota |
| 0.25 | 0.64 | Hood River, Oregon |
| 0.25 | 0.63 | Humboldt, California |
| 0.25 | 0.63 | Albany, New York |
| 0.25 | 0.64 | Bayfield, Wisconsin |
| 0.25 | 0.64 | Marion, Indiana |
| 0.25 | 0.64 | Rock, Wisconsin |
| 0.25 | 0.67 | Lake, Indiana |
| 0.25 | 0.79 | Alameda, California |
| 0.24 | 0.64 | Cumberland, Maine |
| 0.24 | 0.68 | Contra Costa, California |
| 0.24 | 0.83 | Clayton, Georgia |
| 0.24 | 0.62 | Mahoning, Ohio |
| 0.24 | 0.62 | Rock Island, Illinois |
| 0.24 | 0.62 | Hampden, Massachusetts |
| 0.24 | 0.63 | Lane, Oregon |
| 0.24 | 0.63 | Trempealeau, Wisconsin |
| 0.24 | 0.63 | Carlton, Minnesota |
| 0.24 | 0.63 | Cheshire, New Hampshire |
| 0.24 | 0.63 | Grand Isle, Vermont |
| 0.23 | 0.63 | Crawford, Wisconsin |
| 0.23 | 0.63 | Orleans, Vermont |
| 0.23 | 0.63 | Gunnison, Colorado |
| 0.23 | 0.63 | Lackawanna, Pennsylvania |
| 0.23 | 0.63 | Grafton, New Hampshire |
| 0.23 | 0.63 | Portage, Wisconsin |
| 0.23 | 0.63 | Routt, Colorado |
| 0.23 | 0.67 | Broward, Florida |
| 0.23 | 0.65 | Clarke, Georgia |
| 0.23 | 0.62 | Palm Beach, Florida |
| 0.23 | 0.61 | New Haven, Connecticut |
| 0.23 | 0.61 | Eagle, Colorado |
| 0.23 | 0.62 | Howard, Iowa |
| 0.23 | 0.62 | Jackson, Iowa |
| 0.23 | 0.62 | Green, Wisconsin |
| -0.22 | 0.26 | Ector, Texas |
| -0.22 | 0.18 | Kiowa, Kansas |
| -0.22 | 0.18 | Sioux, Iowa |
| -0.22 | 0.18 | Piute, Utah |
| -0.22 | 0.18 | Cleburne, Alabama |
| -0.22 | 0.18 | Brantley, Georgia |
| -0.22 | 0.18 | Campbell, Wyoming |
| -0.22 | 0.24 | Upton, Texas |
| -0.22 | 0.27 | Seward, Kansas |
| -0.22 | 0.16 | Wichita, Kansas |
| -0.22 | 0.25 | Ward, Texas |
| -0.22 | 0.17 | Donley, Texas |
| -0.22 | 0.16 | Morton, Kansas |
| -0.22 | 0.17 | Alfalfa, Oklahoma |
| -0.22 | 0.17 | Holmes, Florida |
| -0.22 | 0.17 | Rock, Nebraska |
| -0.22 | 0.17 | Leslie, Kentucky |
| -0.22 | 0.24 | Winkler, Texas |
| -0.22 | 0.17 | Box Elder, Utah |
| -0.22 | 0.17 | Hooker, Nebraska |
| -0.22 | 0.16 | Jack, Texas |
| -0.22 | 0.17 | Crook, Wyoming |
| -0.22 | 0.17 | Morgan, Utah |
| -0.23 | 0.17 | Bear Lake, Idaho |
| -0.23 | 0.17 | Cullman, Alabama |
| -0.23 | 0.17 | Archer, Texas |
| -0.23 | 0.17 | Caribou, Idaho |
| -0.23 | 0.17 | Sevier, Utah |
| -0.23 | 0.16 | Kingfisher, Oklahoma |
| -0.23 | 0.15 | Hutchinson, Texas |
| -0.23 | 0.37 | Pecos, Texas |
| -0.23 | 0.16 | Jefferson, Idaho |
| -0.23 | 0.16 | George, Mississippi |
| -0.23 | 0.16 | Duchesne, Utah |
| -0.23 | 0.16 | Sterling, Texas |
| -0.23 | 0.16 | Millard, Utah |
| -0.23 | 0.16 | Carter, Montana |
| -0.23 | 0.16 | Roger Mills, Oklahoma |
| -0.23 | 0.16 | Dewey, Oklahoma |
| -0.23 | 0.21 | Garza, Texas |
| -0.24 | 0.16 | Banks, Georgia |
| -0.24 | 0.16 | Sioux, Nebraska |
| -0.24 | 0.16 | Dawson, Georgia |
| -0.24 | 0.16 | Logan, Kansas |
| -0.24 | 0.16 | Cameron, Louisiana |
| -0.24 | 0.16 | Haakon, South Dakota |
| -0.24 | 0.17 | Clark, Idaho |
| -0.24 | 0.14 | Gray, Texas |
| -0.24 | 0.14 | Hemphill, Texas |
| -0.24 | 0.14 | Wheeler, Texas |
| -0.24 | 0.15 | Garfield, Montana |
| -0.24 | 0.15 | Loving, Texas |
| -0.24 | 0.15 | Glascock, Georgia |
| -0.24 | 0.15 | Rich, Utah |
| -0.24 | 0.15 | McPherson, Nebraska |
| -0.25 | 0.27 | Hale, Texas |
| -0.25 | 0.15 | Blount, Alabama |
| -0.25 | 0.15 | Hayes, Nebraska |
| -0.25 | 0.14 | Uintah, Utah |
| -0.25 | 0.15 | Scott, Kansas |
| -0.25 | 0.15 | Arthur, Nebraska |
| -0.25 | 0.15 | Ellis, Oklahoma |
| -0.25 | 0.15 | Major, Oklahoma |
| -0.25 | 0.14 | Sherman, Texas |
| -0.25 | 0.15 | Banner, Nebraska |
| -0.25 | 0.15 | Texas, Oklahoma |
| -0.25 | 0.13 | Hartley, Texas |
| -0.25 | 0.13 | Stevens, Kansas |
| -0.25 | 0.22 | Crane, Texas |
| -0.25 | 0.14 | Blaine, Nebraska |
| -0.25 | 0.14 | Jackson, Kentucky |
| -0.25 | 0.14 | Carson, Texas |
| -0.25 | 0.28 | Dawson, Texas |
| -0.25 | 0.14 | Shackelford, Texas |
| -0.26 | 0.14 | Harper, Oklahoma |
| -0.26 | 0.12 | Lipscomb, Texas |
| -0.26 | 0.12 | Cimarron, Oklahoma |
| -0.26 | 0.24 | Sutton, Texas |
| -0.26 | 0.16 | Gaines, Texas |
| -0.26 | 0.13 | Thomas, Nebraska |
| -0.26 | 0.18 | Martin, Texas |
| -0.27 | 0.13 | Armstrong, Texas |
| -0.27 | 0.13 | Livingston, Louisiana |
| -0.27 | 0.11 | Motley, Texas |
| -0.27 | 0.11 | Oldham, Texas |
| -0.27 | 0.21 | Moore, Texas |
| -0.27 | 0.11 | Borden, Texas |
| -0.28 | 0.12 | Wallace, Kansas |
| -0.28 | 0.17 | Andrews, Texas |
| -0.28 | 0.12 | Franklin, Idaho |
| -0.28 | 0.12 | Madison, Idaho |
| -0.28 | 0.11 | Beaver, Oklahoma |
| -0.28 | 0.11 | Grant, Nebraska |
| -0.29 | 0.11 | Hansford, Texas |
| -0.29 | 0.26 | Deaf Smith, Texas |
| -0.3 | 0.19 | Parmer, Texas |
| -0.3 | 0.09 | Glasscock, Texas |
| -0.31 | 0.2 | Reagan, Texas |
| -0.32 | 0.08 | Roberts, Texas |
| -0.32 | 0.08 | Ochiltree, Texas |
| -0.34 | 0.05 | King, Texas |
* This is 8 years out of date, but by far the most complete and precise data set until the 2010 Census.
Labels: Election 2008
One really interesting thing here is the lowess line itself. Assuming that non-whites’ propensity to vote for Obama is roughly constant among counties, what the results say is that whites living in mixed-race counties are much less likely to vote for Obama than are whites in overwhelmingly white counties. Notice how the Obama vote falls linearly as the mixture of whites increases until the county gets pretty white and then levels off (which can only mean that whites are becoming more likely to vote Obama as you move from about 65% white up). The lowess line actually slopes up at the far right.
Some of this has to be a South effect — there is a big concentration of blacks in the South and whites vote in an ethnic block for Republicans there. It would be interesting to restrict the sample to the non-South and re-plot the lowess line to see if it is just a South effect or if the effect shows up everywhere.
The other thing that would be interesting is smaller geography and individual voting data. Census has demographic info conveniently at the Congressional district level. I wonder if any of the individual level election surveys present geocodes at the congressional district level.
The contours of this map are eerily reminescent of the map you did a couple of weeks ago showing where male life expectancy deviates from income predictors.
Does this mean liberal whites live longer, even controlling for income? (Or liberal states make people live longer…)
If one regards the last resulting shaded map as an approximation of where the most whites voted for Obama as shown in blue and hence are the most liberal, then he most surprising thing to me were the blue strips along the Mexican border in Texas and along the lower Mississippi river valley starting in Northern Louisiana. I suspect both, and especially the later, are an artifact of the method however.
Obama didn’t win the white vote, but neither did any Democrat since LBJ.
In fact, according to this table, Obama did surprisingly well among white voters, for a (black) Democrat.
Supporting Bill above, there does seem to be a Southerner effect: the three best performing democratic candidates were Humphrey, Carter and Clinton – they still lost the white vote, but only by 2-4 points. All were Southerners. The related article notices this as well.
Another interesting parameter to account for would be income, education level and real estate values as proxies for “elites”. Some of the outliers on your %white/%obama list seem explainable by SES stratification among whites.
This could differentiate between plain “whites” and “whiter” people. It could also explain some of the southern white exceptionalism noted.
Humphrey was as un-Southern as you can get. He triggered the 1948 Dixiecrat revolt by winning a civili rights plank at the convention.
Some of this has to be a South effect — there is a big concentration of blacks in the South and whites vote in an ethnic block for Republicans there.
I think this phenomenon may increase as the percentage of minorities in the US increases. I fear that in the future people are going to vote tribally: Hispanics and blacks for the Democrats, and whites for the Republicans.
yeah, isn’t loess awesome? That subtle incline towards the end would be imperceptible in a linear regression or by eyeing the cloud.
But it squares with what people have observed about the *ultra* white districts being less familiar with what minorities are actually like, making them more prone to idealization (and hence Obama voters).
Sounds like the 15% and 50% rule of black politicians. I don’t think its ignorance, but rather that blacks aren’t politically powerful enough to be any sort of threat. Its like people who send their kids to private schools but support bussing. No skin off their nose.