|“Playing with data is as fun as playing with Legos”|
Even though I suspect that most Americans, Republican and Democrat, believe in mostly the same things. The political gap seems to have much to do with your neighborhood – that is to say, how far you live from your neighbors.
I did a little figuring with the population density of each state and the percent of votes cast by its residents for Donald Trump. The correlation coefficient (yes, I’m college educated) was -.46, which apparently is a moderate downhill or negative relationship (see chart #1). In other words, the higher the population density (urban) the less likely you and your neighbors were to vote for Trump. The lower the density (rural), the more likelihood of Trump votes in your neighborhood.
But this gap seems to have been magnified by the U.S. Constitution, as the document describes the Electoral College. North Dakota, 47th in density ranking, cast 216,133 votes for Trump. That amounted to only 72,044 votes for each of the state’s 3 electoral votes for the Republican candidate. In Massachusetts, the 3rd most densely populated state, it took over 100,000 more votes for Clinton (178,615) to earn one of the state’s 7 electoral votes for the Democrat (see chart #2).
What surprises and disturbs me is the education gap. The graph below, from Pew Research Center, indicates that among all voters, those with college degrees or more voted for Hillary Clinton by 9 points, while voters with some college or less chose Donald Trump by 8 points. The education gap widens when looking at white voters only, a gap of 35 points.1
There are many ways to read meaning into this, and I’m going to be thinking pretty hard about it. But we might assume that free college education, as provided in many European countries, is pretty much off the table here at home.
I woke up early again this morning, all worried about this upcoming election. I started mucking around my old 2009 Macbook Pro and found the Federal Elections Commission web site and their downloadable files with details on campaign contributors by state. Data makes my skin tingle.
So I downloaded all 27 megabytes of North Carolina data (4/15/15-10/31/15), loaded the csv file into Open Office Calc and started tinkering. My seven year old MacBook was huffin’ and puffin’.
One of the questions that got my mind going this morning was the money that is so essential to political campaigns today. To date, the 2016 presidential campaigns have generated $1,000,058,201 from individual donations alone. More to the point of my sleeplessness was, “Who’s paying for these campaigns?” or “Who’s buying our government?”
So I used Calc to parse the 133,100 contributions by range categories: less than $100, $100 to $999 and more than $1000 and more. It shouldn’t be a surprise that more North Carolinians were donating less than $100 than the other two combined.
What struck me as especially critical to my worries was the total amounts of campaign money generated from each category. Look at the data and graph.
|Donations||Number of Contributors||Total Amount Contributed|
|Less than $100||101,388||$2,737,190.87|
|Between $100 & $1000||28,427||$5,454,833.10|
|More than $1000||3,285||$6,226,996.52|
What’s wrong with this picture? Well, let’s say you are an incumbent, or even a challenger. With so much money out there, constituting a elections industry, the only way that you can keep your seat, or oust the incumbant is with a lot of money.
Where do you go for the money?
Look at the diagram again. Where’s the money? To get elected, you have to convince rich people and corporations to contribute. What will they want from you for that money?
It’s their government. Not ours.