Twitter in the Midterms

Whereas virtually all candidates involved in competitive races for the House of Representatives in this year’s midterm elections were on Facebook (see previous post), Twitter use was lower, as one would expect from a newer and less widely-used platform. On November 2, 110 out of 112 Republicans running in competitive districts were on Twitter, 92 out of 112 Democrats.

Many candidates on Twitter are met with deafening indifference. The median number of followers of Republican candidates is 494, of Democrats, just 255. On Election Day, 19 out of the 92 Democrats and 2 out of 110 Republicans using this tool belonged to the estimated 97% of Twitter users who have less than a 100 followers. Ten-term incumbent Rep. Earl Pomeroy (D-ND) had 6 followers on the night he lost his seat to State Rep. Rick Berg. By comparison, President Barack Obama had almost 6 million Twitter followers in early November 2001, and former Alaska Governor Sarah Palin about 300,000. If we want to understand the use of Twitter in politics, we have to focus on the median candidates and the Earl Pomeroys of this world, and not only on successful outliers like Obama and Palin.

The figure below shows the distribution of Twitter followers, from data Cristian Vaccari and I have collected (see note below on the sample of districts).

As with Facebook supporters, the curve is steep—on Election Day, Jesse Kelly (R-AZ) had 13,726 followers, 11% of all the Republican followers, and Alan Grayson (D-FL) had 11,573 followers, 21% of all the Democratic followers in the dataset. (Both lost, incidentally). Grayson is no shock here, but in the light of her extraordinary Facebook support (139,203 on November 2) I was surprised to see Michele Bachmann a distant third amongst the Republicans, with 8988 followers.

The next figure compares the Twitter following (dotted lines) with Facebook supporters (thin solid lines) for the two parties. (It’s cropped to leave out numbers over 14,000, which means that three Republicans’ and one Democrat’s Facebook support does not appear here.)

Unsurprisingly, the numbers for Twitter are generally lower than those for Facebook (though some individual candidates, like the above-mentioned Jesse Kelly, have more Twitter followers than Facebook supporters). The Republican average is 1186, the median, as said, 494. For the Democrats, the average is 600, the median 255. The Facebook numbers are, by comparison, 4223 (Republican average), 2156 (Republican median), 2149 (Democratic average), 1609 (Democratic median).

I think these figures document the limited reach of most politicians’ use of Twitter, which is an important point to keep in mind in discussions of social media’s role in politics—this should not, however, be taken to mean that this microblogging platform is politically irrelevant as such. First, as noted from the outset, there are a few politicians who have build very considerable Twitter followings. Secondly, much politics does not directly involve politicians, but simply citizens talking and acting together, and tools like Twitter and Facebook have a role in that too, though it is a role not considered here in this post.

Note: The data comes from major party candidates running in any of the 112 districts rated as lean Republican, tossup, or lean Democrat by either the New York Times, the Congressional Quarterly, Cook, or Real Clear Politics in August, 2010.