In my last post on voter turnout, I promised an update when final numbers were in…and here we are.  In the meantime, I’ve been researching the topic and found Dr. Michael McDonald’s United States Elections Project to be extremely valuable.  I’ll summarize, but I encourage you to check out the US Elections Project Web site.

The voting-age population (VAP) is traditionally used to calculate turnout, and includes anyone of voting age living in the US.  However, as Dr. McDonald argues, the VAP includes non-US citizens and felons who are ineligible to vote, while it excludes eligible voters living outside of the US.  He has come up with a way to calculate the voting-eligible population (VEP), which attempts to correct for these flaws.  I find the VEP much more precise for determining voter turnout.

The fascinating thing about his research is that it reveals that voter turnout has not been as low in recent decades as previously thought.  Check out his chart below, which shows the difference in voter turnout calculations between VAP & VEP:

Presidential Turnout Rates: 1948-2008, from the United States Election Project

Note the ever-widening gap between VAP turnout rate (yellow) and VEP turnout rate (green).  Since 1972, the VAP has become less accurate for determining voter turnout; almost a 5% difference in 2008.

How’s that for a prelude?  Now we’re ready to take a look at this year’s election turnout, which was 61.6% based on VEP calculations; right up there with the 1968 election.

I took Dr. McDonald’s 2008 state election turnout rates and mapped them using Inkscape.  The map is in Scalable Vector Graphics (SVG) format (Firefox, Opera & others should be fine, but IE users will need the Adobe SVG viewer plugin).  Unfortunately SVG is not a standard yet, but based on my limited exposure to it so far, I’m a fan.  In Firefox, hovering over a state will reveal the % voter turnout for that state (I’m not sure why I can’t get this to work with Opera & IE).  Select the thumbnail below for the SVG version (or if you prefer, here’s a link to the less-cool but stable JPG version).

Voter Turnout by State in the 2008 Presidential Elections

And below are my findings & analysis.  Standard JPG format here.  Of particular interest to me was the fact that the blue states had 4.9% higher turnout than red states in the 2008 election.  It was also fascinating that Minnesota has the highest turnout for the past 3 presidential elections.  I wonder if that hotly contested Senate race had anything to do with it this year…  As usual, select the image for a larger version.

Summary of Voter Participation in the 2008 Presidential Election

Thanks again to Dr. McDonald for his excellent work on voter turnout.

UPDATE 11/7/2008: The Federal Election Commission has lowered it’s preliminary estimate of people who voted in the 2008 election to 122,253,622.  That’s only 61.9% of all eligible voters, which hasn’t been bested since 1964 (Lyndon Johnson over Barry Goldwater).  But the numbers remain preliminary, so I’ll revisit everything when it is finalized.  Thanks!  –frets

***

Yesterday’s election drew voters out in record numbers, if you couldn’t tell by the lines some of you experienced at your polling places.  I voted early and had to wait about 40 minutes, which isn’t so bad…but I’m not in a swing state!

Based on preliminary results, 148,218,161 million people voted.  That’s 64.1% of all eligible voters.  It may not seem like much, but that’s one for the record books, especially in these times.  That percentage has not been exceeded in our lifetime (unless you’re a hundred years old)!  You have to go back to the election of 1908 to beat it!  Back then, William Howard Taft (R-Ohio) beat William Jennings Bryan (D-Nebraska) with 321 electoral votes, and 65.4% of eligible voters cast a ballot.  Incidentally, the all-time record goes to the hotly-contested race between Rutherford B. Hayes (R-Ohio) and Samuel J. Tilden (D-New York).  Hayes eeked out a 1-point electoral victory, and 81.8% of eligible voters cast a ballot!  Talk about long lines!

Here’s a look at the voter turnout since 1960 (select the image for full-size):  Read the rest of this entry »

Ever since my last post, I’ve watched the gas price at local stations keep coming down!  What a welcome relief.  It seems like the only thing coming down faster is the stock market (ba dum bum).  But seriously folks, I wanted to see how this year stacks up against prior years, so here it is:

U.S. Gas Prices Leading up to the 2008 Election

Now THAT’s a clear downward trend prior to the elections!  There are some plausible explanations for this dramatic (possibly record-setting) drop in gas prices (other than manipulation by the powers that be).  And without corroborating evidence from prior years, I’m inclined to believe them.  Besides, President Bush already demonstrated his office’s inability to influence crude prices & supply.

Are gas prices manipulated to go down prior to a Presidential election?  My research led me to look at the national level because the data for states is difficult to ascertain, and probably irrelevant.  According to the EIA, in particular their Oil Market Basics publication, the largest variance in gas prices would occur at the regional level, largely due to supply issues.  Looking at the breakdown of what makes up the price of gas, we get the following:

Distribution and marketing are mostly the profit the local station makes.  Taxes, then, are the only portion of cost over which a state has control.  Taxes are relatively steady and typically don’t change month-to-month.  So what made sense was to look at the national level, to see if there were trends there (select the image to see it bigger):

Gas Prices and Presidential Elections, 1992 - 2007

No smoking gun here.  If anything, there seems to be a trend in recent non-Presidential election years for gas to drop prior to November (2001, 2003, 2005, and 2006).  But in contrast, gas prices in the Presidential election years have remained relatively flat or, in the case of 2004, have increased prior to Election Day.  If there were some manipulation going on, I would have expected to see a stark contrast in the trends of election and non-election years.

I have to admit, I wasn’t sure how this one would turn out; but it appears that this myth is busted.

Source: Energy Information Administration

Now that the Democratic and Republican National Conventions are over, I grabbed all of the (available) speech texts from the DNC and RNC Web sites and threw them into wordle.net just for fun (yep, I’m sadistic that way sometimes).  It is hard to compare word occurrences because the Republicans don’t have nearly as many speeches available (52) as the Democrats (168).  The Reps lost a day to Gustav, but they also seem to only have the major speeches available online, whereas the Dems have everything available.

Some interesting findings in the data (hover for definitions):

The Obama-to-McCain ratio (DNC): 2.9:1
The McCain-to-Obama ratio (RNC): 8:1
The Obama-to-Bush ratio (DNC): 4.9:1
The McCain-to-Bush ratio (RNC): 88:1 Bush was only mentioned 4 times in speeches available at the RNC.
The Obama-to-Biden ratio (DNC): 9.9:1
The McCain-to-Palin ratio (RNC): 8:1

Here are the word clouds:

All DNC 2008 speeches word cloud

All RNC 2008 speeches word cloud

Here’s the top 15 words:

Here are links to the word clouds at wordle.net:

And if you haven’t had enough, I’ve linked both the DNC speech word counts and RNC speech word counts as text (csv) files.  Enjoy!