Election 2012 Statistics

(November 6th) Being someone who is, by now, completely sick of hearing about politics, I find my self in the surprising situation of wanting to talk about politics.  Strange, I know.  First off, big caveat, no ideologies in this post, just some interesting statistics.  I came across this article in the Washington Post, you should give it a quick perusal if you can, that presented the exit poll statistics as functions of gender, age, party, race, education, income, gender x marital status, ideology, religion, evangelicalism, financial situations, top issues, top qualities, and national economic condition perception.  It's surprisingly easy to glance over and gleam a lot of information from, which is why I'm dedicating a post to it here.  Certainly worth the two minutes it takes to look it over quickly.  What really struck me wasn't that Obama lost some percentage points across a large number of groups (these are small changes that reflect that he won by a smaller margin as compared to four years ago), but rather the trends related to evangelicalism and economics:

Screen shot from the Washington Post.  Graphic by Sisi Wei.  Published online Nov. 6th, 2012.

The way to read the graph is that where the arrow points indicates the number of percentage points above the other candidate that that candidate scored for that group of the population.  The easy way to think of this is that for a candidate, let's say Obama, the left end is 100% of the population voted for him, the middle line is 50%, and the right end is 0%.  It'd be the opposite for Romney obviously.  The arrow length indicates the change from four years ago.

What I want to bring attention to in the top plot is 1) the sharp divide between evangelicals and all others, and 2) the dependence of a person's financial situation on the way they voted.  For those whose financial situation improved or stayed the same, they were more likely to vote for Obama than the other way around.  This is also seen in a different way in the next plot:

Screen shot from the Washington Post.  Graphic by Sisi Wei.  Published online Nov. 6th, 2012.

From the above plot, we learn that four years ago, people who thought that the national economic condition wasn't good (let's rephrase this as not improving) tended to vote for Obama over McCain, and now, people who think that way are more likely to vote for Romney.  Makes sense...if the country isn't doing well in your opinion, then you'd like to see change.  The surprising thing about how the change is manifested for people who think the country is doing well (or in rephrased terms, that the country is improving) is the magnitude of the swing...from four years ago the break down was approximately 75% of the people who thought the country was well off economically voted for McCain, now some 90% of the people who think the country is well off voted for Obama.  I'm not going to pass judgement on why one would think the country is well off, but the magnitude of that swing is very indicative of the polarization that we see today in our liberal/conservative dichotomy.

The very last thing to bring to one's attention is that people who think that the top quality is that a president cares about you tend to vote Democrat (by about a 3:1 margin).  Those that think the top quality in a presidents is that they are a strong leader, shares my values, or has a vision for the future tend to vote Republican, but by a more even margin.

Anyways, thought this was interesting.  Click the link if you want the rest of it.