Wednesday, October 24, 2012

Mapping the Election

My last post about the history of U.S. political parties included a number of charts detailing a history of change over time in voter support for third party presidential candidates. Historians, like myself, tend to handle "change over time" well. But where we are often less skilled is in making sense of spatial change in contrast to, or even in addition to, temporal change. It is easy enough to explain that things used to be one way, then some time passed, and now things are a little bit different. And, of course, historians not only produce these kinds of narratives of "change over time" but also provide an analytical interpretation of why those changes occurred.

But after posting last week's essay and accompanying figures, I realized that my analysis was missing one big component. I got "change over time" pretty well, but I totally neglected to discuss "change over space."

Therefore, below is a bit more analysis to accompany my earlier post. After reading the first post, come back here and check this out. What I have done is map third party presidential candidate support not over time, but over space. And what I found is that space matters a lot in U.S. politics. To put it another way, in presidential elections states matter a lot more than the federal or national level does. This is partly because of the Electoral College system, and partly because of extremely complicated and sometimes quite arcane ballot access laws within each state. We already know that the race for the presidency is a race to win just a few key "swing states." But what I did not know until I looked more closely at the data was that third party candidates are also racing for states as well.

The Geography of Third Party Presidential Support

Percentage of Voters Selecting Third Party Presidential Candidates, on a scale from 0% to 4%,
U.S. Presidential Election, 2008 (Source: I made this!)

The above map graphs the intensity of voter support for third party and independent presidential candidates in each state for the 2008 presidential election. I chose to analyze the data from 2008 because it wasn't an especially good year for third parties (although all three parties I have been analyzing in this study — Green, Libertarian, and Constitution — did better in 2008 than they did in 2004) and also because it was the most recent U.S. presidential election. Therefore the data is (for the most part) not skewed by the presence of a "celebrity" third party candidate (but we'll have to deal with Nader as an independent candidate) and also is hopefully close to the way American voters still feel about third parties in 2012.

The intensity of voter support for third party presidential candidates in 2008 ranged from a high of 3.2% of voters in Montana to a low of 0% of voters in Oklahoma. (It seems that Oklahoma did not allow a single third party candidate onto its ballot in 2008. So is Oklahoma the center of anti-democracy in this country? I actually have no idea why their ballot rules are so strict; something to explore...) Overall, the map makes clear that western states (including Alaska) were generally much more willing in 2008 to vote third party than eastern states were, with the exceptions of a few interesting outliers in New England.

Before I made this map, I considered this hypothesis: voters in "swing states" are probably less willing to vote third party, whereas voters in "solidly red" or "solidly blue" states are probably more willing to give their vote to other parties and candidacies. But this map has largely proved that hypothesis wrong. Sure, it should be no surprise that Montana and Utah voters gave over 3% of their votes to third party candidates in 2008; they are solidly "red" (Republican) states and voters knew that Obama had no chance of winning their state. Same thing could be said, perhaps, for "solidly blue" (Democratic) states like Maine, Massachusetts, and Vermont. 

But when you look at the "swing states," the data is a bit inconsistent. For example, over 1.5% of Ohio voters supported third party candidates in 2008, more so than any of the "safe" or "safer" states around Ohio. However, in other "swing states," like Florida (0.75% third party support) and Virginia (1% third party support), the hypothesis seems to have held true. But then I look at my home state of New York, a "solidly blue" state, and see that just 1.1% of voters supported third party candidates. Why did so few New Yorkers vote third party when they knew that Obama was going to win? So, in the end, the hypothesis of "swing states" vs. "safe states" doesn't hold up. Some states prove the hypothesis; others do not. So we must throw it out. People didn't choose to vote or not vote third party in 2008 because of the "redness" or "blueness" of their state. There must have been other factors involved.

In fact, while it seems to be generally true that western states and some parts of New England are more "independent" than other states in the nation, I think each state has its own specific reasons for why it either goes or does not go third party. These may include: ballot access laws, third party support at the local/regional/state level within each state, swing state/safe-state status, hometown heroes and allegiances among the candidates and the places they hail from, et cetera.

For example, just look at Montana. It looks like Montana had the highest third party presidential support of any state in 2008. But in fact, over 2% of the total 3.2% of Montanans who went for third parties in 2008 wrote in Ron Paul for President even though Dr. Paul was not actively campaigning for the office. Yup, that's right. More people in Montana voted for Ron Paul (who wasn't running, and voters had to write-in his name) than voted for Ralph Nader plus all three major third party candidates combined!!

So, to figure this situation out, I determined that it would be helpful to see how each third party candidate performed, state by state, in the 2008 race:

Percentage of Voters Selecting Libertarian Party Candidate Bob Barr, on a scale from 0% to 1%,
U.S. Presidential Election, 2008 (Source: I made this!)

Bob Barr, Libertarian Party candidate for President in 2008, did a good job in 2008. He did not achieve the best showing ever for a Libertarian candidate, but he performed better than all other third party candidates that year (and only slightly worse than Ralph Nader who ran as an independent). He won about 0.4% of the national popular vote. But as you can see from the map above, his support varied across the country from state to state.

Barr's best performance was in Indiana where he received just over 1% of the total vote. It is not clear why Barr did so well in Indiana. He seems to have had no personal relationship with the state. But, on the other hand, it seems that he was the only third party or independent candidate to achieve ballot access in Indiana. By virtue of that, almost all votes placed against Obama and McCain therefore went to Barr. Barr also did well in Georgia (his home state), where he received about 3/4 of one percent of the vote. And he also received the same share of the vote in Utah and in Texas. As you can see, Barr's worst performances were generally in the east, especially the Deep South and the Northeast. (Also, gray states represent those states where Barr failed to achieve ballot access.)

The fun thing about this map will be to compare it to Gary Johnson's performance two weeks from now in the 2012 election. Most people predict that Gov. Johnson will win much more of the national vote than Barr did four years ago. But where will his strong states and weak states be? Will we see a pattern in Libertarian support across the country from 2008 to 2012? Or will Johnson's results look completely different than Barr's?

Percentage of Voters Selecting Constitution Party Candidate Chuck Baldwin, on a scale from 0% to 1%,
U.S. Presidential Election, 2008 (Source: I made this!)

Constitution Party candidate Chuck Baldwin won only 0.15% of the national popular vote in 2008. Compared to Barr's performance for the Libertarian Party (above), it is clear from this map that Baldwin's performance was more scattered. He did well in some states but also very poorly in others. 

Baldwin's best performance was in Utah, the only state in which he won over 1% of the vote. He won 1.26% of Utah voters' support, better than Barr performed in any state, in fact. Beyond Utah, Baldwin's best performances were in Idaho, Alaska, and South Dakota. He performed very poorly, however, in many large states like California and New York, and he generally performed poorly in the South and in the Northeast. Like other third party candidacies, the Constitution Party performed best, generally, in the west. I wonder if Mormon support wasn't perhaps critical in states like Utah and Idaho for providing Baldwin with his best showings.

This year the Constitution Party candidate is Virgil Goode of Virginia. He is expected to do better than Baldwin did in 2008. But where will he find support? As the only southerner on a major ticket for the Presidency, will he do particularly well in the south? Or will the west still be his best shot at glory? Of course Chuck Baldwin was also a southerner (if you count Florida) but he didn't do well there. So native place might, after all, not always count for much in third party politics.

Percentage of Voters Selecting Green Party Candidate Cynthia McKinney, on a scale from 0% to 1%,
U.S. Presidential Election, 2008 (Source: I made this!)

Green Party candidate Cynthia McKinney did not do very well in 2008. She received only 0.12% of the national popular vote. The Greens only got on the ballot in 32 states, limiting their chances to receive more votes nationwide. Their struggle with state-by-state ballot access laws as well as weak support elsewhere (perhaps due to Ralph Nader's independent candidacy) resulted in a scattering of support across the country, as seen in the map above.

McKinney's best performance was in Louisiana (of all places) where she received just under 1/2 of one percent of the vote. She did much better than all other third party candidates in Louisiana. I never thought of Louisiana having a very active Green Party, but perhaps they do. Or perhaps McKinney focused much of her campaign effort in that state. I know that bringing attention to the government's failures to help black residents of New Orleans after Hurricane Katrina was an important issue at that time, and she may have won support among Louisiana's African-American population in 2008. McKinney also did well in Maine (0.4% of the vote). It is known that Maine has the highest percentage of enrolled Greens of any state in the nation, so the Green Party could have expected to do well there. But elsewhere, support for her candidacy was lukewarm, and as can be seen, in nearly twenty states in the middle of the country she was not even on the ballot.

This year's Green Party candidate, Jill Stein, is expected to perform better than McKinney did in 2008. But it is not clear where she will find the most support. She has achieved ballot access in many more states than her predecessor did, so that will help. It will be interesting to see if perhaps the Greens buck the trend of other third parties and find more of their support in the eastern rather than the western states. Or perhaps we will find that all third parties do better in the west than in the east, perhaps because people there are just more independent thinking? Let's wait and see.

Top Third Party Presidential Candidates per State, shaded by their percentage of that state's vote, on a scale from 0% to 1%, 
U.S. Presidential Election, 2008 (Source: I made this!)

Lastly, I thought I'd share this final map. The map above shows which of three major third parties (Constitution [purple], Green [green], and Libertarian [yellow]) performed best in each state, as well as showing the overall level of voter support within each state for that candidate on a scale from 0% to 1%. It is an interesting map, but mostly because I look forward to comparing it with data from the 2012 election just two weeks away.

You can see that in 2008 the Libertarian Party candidate performed best in most states of the union. Libertarian performance was strongest in the west, as mentioned, but also in Georgia (the candidate's home state) and in Indiana. The Constitution Party did best in at least eight states: Alaska, Oregon, Idaho, Utah, Nebraska, Mississippi, West Virginia, and Connecticut. The party did best, as mentioned, in Utah and Idaho, and just barely bested the Libertarians in the other six states. Meanwhile, the Green Party only performed best in two states (Louisiana and Maine) and in one district (the District of Columbia — perhaps you can see the little green dot I put on the map!). 

If you left this to my imagination, I would have imagined the Greens sweeping the Northeast and the West coast, while Libertarians would win all the rest of the west, and the Constitution Party would excel in the South. But that's not exactly what these maps have demonstrated. We know that things are much, much more complicated. And that complication should bleed over into the way we view "red" versus "blue" states. These maps get us to think beyond the simple binary of "red" versus "blue." What about "green states," "yellow states," "purple states?" we might ask. Yes, they also exist. Or at least, as of right now, we have red states with hints of yellow and blue states with hints of green and all kinds of other beautiful combinations of colors. In two weeks, our country is not just going to decide whether we want a "red" or "blue" president. We are deciding, in each of fifty different states, whether we want a red, blue, purple, green, yellow, or other-colored president. You see, democracy is all about choices, and our country's democracy will become stronger only when we start celebrating the rainbow of political perspectives, platforms, parties, and candidacies that exist, and giving each color its fair share, rather than continuing to see the world in black and white.

No comments:

Post a Comment