Saturday, November 6, 2010

Is Voter Turnout in NC Lower or Higher than the National Average? Yes

There was a lot of pre-election speculation about voter turnout--in North Carolina and nationwide.  We're finally starting to get some reliable data and it appears that turnout was up from 2006--the last midterm election.  In North Carolina, the proportion of eligible voters* who turned out was up about 8% (to 39.7%) compared to a voting eligible turnout rate of 41.5 nationally (up a bit from 40.4%).

Because I prefer pictures to words, I've pasted a graph of midterm turnout from 1982-2010.  The blue line represents turnout in NC (an odd choice given the recent legislative turnover) and the red-line is U.S. turnout.

In addition to the large spike from 2006 to 2010, there are a couple of other interesting stories here.  The most obvious one is how spikey the NC line is, compared to the comparatively flatter line nationally.  Clearly there's something that makes turnout in North Carolina vary considerably from one midterm to the next.  My guess is that it's higher in years with US Senate elections, but I have not checked it to be sure.

So what happens in Presidential election years?  Below I've made a similar graph, but this one just includes years in which we're voting for President.  Here we see a slightly different story. Turnout in the Tar Heel state was consistently below the US averages for a long time, but has recently shot above the US average.  I would say this is simply an Obama effect, but the rise in NC actually began in 2000 (when Barack Obama was a state senator in Illinois' 11th district).  We know that competitiveness breeds turnout so my guess is that increased turnout in North Carolina in Presidential election years has to do with the increasing competitiveness of Presidential elections in North Carolina. 

So if you're in a debate with your friends about whether North Carolinians are more likely to turnout than people from other states (likely scenario, I know), you can safely take either side of the argument.  If you're taking about Presidential elections, the answer is yes (or at least it was in 2008).  If you're talking about the midterms, the answer is no.

As more data become available, I expect to post more (including some analysis of what happened in the local elections).

*You may note that these numbers look a little higher than those reported elsewhere.  That's because they measure the proportion of the electorate who did vote divided by the proportion who could vote (the eligible electorate).  The traditional way to measure this was with the voting age population as the denominator.  The problem is that there are lots and lots of people (96,404 in North Carolina) who are above the age of of 18, but who are ineligible to vote (because they are on probation, on parole, in prison, or are not citizens).  Including them in voter turnout numbers makes turnout appear lower than it actually is.  If this is confusing, fret not, I'll probably write a more extended post on this soon (with graphs, of course).

No comments:

Post a Comment