March 5, 2010
Posted by David Karpf
“Republicans rule twitter?” Eh, not so fast…
The Christian Science Monitor is running a story today about a new Congressional Research Service study showing that more Republican Members of Congress (MoCs) are signed up for Twitter than their Democratic counterparts. The story’s headline, “Social Media Domination: Republicans rule Twitter,” has led to some furious retweeting from the political left and the political right and is quickly getting blown out of proportion.
Let me start by noting that I see nothing wrong with the study itself. The CRS counted the total number of MoCs on twitter in September and October ’09, counted total tweets, categorized by content-type, and packaged it all into reader-friendly graphs. It’s exactly what the CRS is *supposed* to do, and I’m not objecting to it.
But the media framing is a little bit silly. Consider this passage in the Christian Science Monitor piece:
“It’s the House that’s the Twitter GOP hotbed. Fully half of the Capitol Hill Twitterverse is composed of House Republicans. Obviously they’ve got some organized Twitter strategy going on in the GOP caucus. (emphasis added)
Is it only Decoder that finds this counterintuitive? It’s Democrats who are the party of young people (who text a lot), and Change, with a capital “C,” and MoveOn, and Web-based fundraising, and so forth.”
The problem here is that “get more MoCs to tweet” isn’t a strategy, per se. Generally speaking, strategy involves making choices about the mobilization of resources to accomplish, y’know, a goal. Furthermore, the Democratic coalition does include lots of young people, MoveOn, web-based fundraisers, etc. But the success of those actors has very little to do with Tom Perriello’s (D-VA) twitter activity. CRS isn’t looking at the Republican and Democratic coalitions, it’s looking at Republican and Democrat Congressmembers. Those Congressmembers, on average, don’t even have a very large following (the median House member had 1,297 followers, the median Senator had 3,536 followers), and we have no information on how frequently they are actually interacting with their followers. Bottom line: if more Republican elected officials are contributing content in 140-character bursts… so what?
This isn’t to dismiss Twitter’s utility in the broader social media universe. I’ve been amazed at how twitter has evolved in the past year and a half. It’s now an important means of directing traffic to blogs, with an astonishing clickthrough rate. But Twitter is an ecology in which communities of practitioners can interact and spread information. Raw numbers of accounts or tweet totals just aren’t very interesting or useful data. In the rush to embrace the newest and shiniest of the social media tools, twitter is also a bubble, being misapplied in areas where 140-character bursts just aren’t all that useful.
*If* Republicans actually rule Twitter, we should see that show up in meaningful action rates. Show me twitter-directed donation numbers, or twitter-mediated protest activities. Those are areas that have genuine strategic importance.
It’s entirely possible that Republicans do lead in these areas, but we can’t tell from the CRS study. That wasn’t its purpose or design. In the meantime, it would be great if journalists and the twitterati would tone down the rhetoric a bit. We have no evidence of Republicans “dominating” or “ruling” twitter. There are just more Republican Congressmembers using the microblogging tool. That’s a little interesting, but hardly newsworthy.
4 Comments
March 5, 2010
This isn’t to dismiss Twitter’s utility in the broader social media universe. I’ve been amazed at how twitter has evolved in the past year and a half. It’s now an important means of directing traffic to blogs, with an astonishing clickthrough rate. But Twitter is an ecology in which communities of practitioners can interact and spread information.
Even more, Twitter is part of a larger ecology that both makes it more important (as you note with click-through rates to other resources) and potentially less important (if it’s just Rep. Whackadoodle grumbling about something) than other channels and aspects of social media.
Twitter can be great but this tendency to hype [SHINY NEW THING] without actually examining metrics is one of my least favorite aspects of politica/tech/all journalism.
March 5, 2010
CRS is lazy or dumb if they didn’t count follows/followers. Also, how many @’s to they give and get. These would be real engagement metrics.
Is the point of the study to express how far the whole lot of them are from Gettingittown, USA.
June 12, 2010
This recent empirical study of Twitter use by Members of Congress may be of interest: Golbeck et al. on Twitter Use by the U.S. Congress (2010), http://j.mp/djZbL6
June 17, 2010
Thanks for the link, Robert. I’ll take a look at it. Are they finding much that is at odds with the CRS study?
In general, I think the greatest potential of a new information architecture like twitter comes from how it gets adopted by communities-of-interest, rather than how it gets adopted by Members of Congress and other established elites. If history is any guide, we’re going to see MoC’s mostly using the new information architecture in the same way they used the old (distributing press releases, credit-claiming, position-taking, etc) , with one or two outliers who experiment on the medium (and those outliers are usually backbenchers with little at risk). Same goes for the established advocacy groups.
Meanwhile, challenger candidates with little to lose and upstart advocacy communities/groups forge ahead with the medium, hoping to establish a foothold that will rapidly expand their otherwise-fledgling efforts. And it works for a couple, which then leads to broader adoption. (in other words, it’s a disruptive innovation story, but on the political instead of economic stage)
As such, I’m just not sure how much the empirical study of Congressional tweets can tell us about twitter’s broader impacts. It’s nice to have the data as a baseline for later work, but I’d worry about drawing any conclusions about the importance of the medium based on that sort of sample.
(then again, that’s just me…)
Sorry, comments for this entry are closed at this time.