:::: MENU ::::
Browsing posts in: Research

U.S. Congressional Mention Networks

I used the Twitter Database Server and my own Twitter-collectors to gather 42,813 tweets posted by 417 elected members of Congress (69 Senators and 348 Representatives) between December 22, 2011 and March 15, 2012. From those tweets, I made a network based on when Members of Congress mentioned each other and used NodeXL to analyze and graph the data. Here are some preliminary results from that network of 2613 mentions.

First, a graph of the whole network. Here, the nodes are colored by their parties (blue for Democrat, red for Republican, yellow for independent), and the edges are colored by whether the source and target are in the same target (blue or red) or if the source and target are from different parties (yellow). The first thing I notice is how different this graph looks from Adamic and Glance’s “Divided They Blog” graph. The divide between parties is less extreme in the Congressional mentions network. Neither graph though, bloggers or MOCs, can tell us whether connections are attacks or endorsements. I think that on Twitter a cross-party mention is likely an attack, but I need to dig into the text of tweets to be sure. If I’m right though, and cross-party mentions are attacks, I find it interesting that Democrats are more likely to mention Republicans than the other way around (at least in this dataset). I think being the minority party in the House drives much of that – i.e. House Democrats are criticizing the House Republican leadership. Again, I need more data to be sure.

I’m working on a paper with Jahna Otterbacher and Matt Shapiro to answer some of these questions and to better understand how Congressional Twitter networks reflect and differ from their offline networks. For instance, do lower-influence MOCs (by offline measures like bill cosponsorship and seniority) garner more influence online than they do on the Hill?  Can social media networks predict alliances or other legislative behaviors? We’re also analyzing what Congress is talking about so we can answer questions such as “Do social media conversations impact the legislative agenda?” and “Do MOCs forefront different issues in social media than they do in traditional media?” All of these questions are part of my broader goal to understand relationships between social media and social action. Baby steps. Stay tuned.

Click for bigger version at NodeXL graph gallery

One-mode Projections in NodeXL

Many of the networks I retrieve and analysis are two-mode affiliation networks – meaning they’re networks of people connected to share objects or communities. For instance, Senators using the #gopconference hashtag on Twitter, where I’ll have an edgelist of Senators and the hashtags they use. NodeXL, my free network analysis software of choice, doesn’t currently support two-mode data though, and one-mode projections of affiliation networks are pretty interesting anyway. So, I worked with a Freelancer to write a macro to produce one-mode projections of two-mode affiliation networks within Excel. You can find the macro as a text file at Github. Continue Reading

Public Officials and Social Media talks at MPSA

Next week the Midwest Political Science Association meets in Chicago, and I’ll be presenting two papers. Here’s my schedule (locations TBD), abstracts after the jump:

Thursday, April 12, 8:30am

Social Media and Other Web-based Networks

Doing What I Say: Connecting Congressional Social Media Behavior and Congressional Voting

Friday, April 13, 10:25am

State, Urban and Local Political Networks

Chicago Politicians on Twitter

Continue Reading

Hiring and Placement in the iSchools

A growing body of literature examines trends in department prestige, graduate employment, and faculty hiring in academic fields such as communication, computer science, economics, higher education administration, and political science. Until recently, we had no similar empirical literature about the relative prestige or reputations of information graduate programs. Emilee Rader (now at Michigan State) and I collected data about graduate placement and faculty hiring in iSchools between 2004 and 2010, and using that data, developed two ranking mechanisms for information graduate programs.

In summary, our data indicate that

  • 14% of iSchool graduates are placed in tenure-track positions at iSchools
  • 40% of iSchool graduates are placed outside academia
  • Less than 50% of the tenure-track faculty hired by iSchools graduated from iSchools

Continue Reading

Qualitatively Coding Tweets

In studying politicians on Twitter, one of my goals is to understand what they’re talking about. The trouble is, tweets are incredibly difficult to code. Researchers at Maryland claimed success with a coding scheme for Congress’ tweets, but my colleagues, students, and I were never able to reach acceptable inter-rater reliability using their scheme (see our new scheme after the jump). We tried a few times, even met to discuss and adjust disagreements, and now I’m suspicious about the reliability of Golbeck’s scheme. The authors don’t provide their kappas, just percent agreement. The problem there is that percent agreement isn’t a good measure of reliability. Especially when the categories are numerous, broad, or incredibly narrow, high percent agreement can be misleading. Matthew Lombard has an excellent guide to interrater reliability where you can learn more. Continue Reading

ChiBudget dominates Twitter discussion too

Chicago Aldermen Joe Moreno (Ward 1) and Brendan Reilly (Ward 42) live tweeted the Chicago budget meetings in late October. As the visualization below shows (click it to see the big version at Many Eyes), the #chibudget hash tag dominated all Aldermen’s discussions between 10/24 and 11/4. Even though 28 Aldermen have Twitter accounts, only 19 posted during that period. As a social media junkie and progressive, I’m glad to live in Ward 1 with Alderman Moreno on my side.

Aldermen Tweeting Many Eyes

A Little Press, Some Acceptance for Public Officials on Twitter Projects

The Aldermen and Congress on Twitter projects made it into popular press and another conference this morning. You can read the popular press story from the Medill News site and conference abstracts below the jump. The papers investigate connections Aldermen make with their constituents via Twitter and how the language members of Congress use can be used to predict their offline political behaviors. Continue Reading

Together they Tweet?

In 2004, Lada Adamic and Natalie Glance published Divided They Blog, a paper in which they report a stark divide between left- and right-wing bloggers. They found relatively few links between liberal and conservative bloggers and more links among conservative bloggers than among their liberal counterparts. I asked whether Congress’s online conversations reveal a similar divide. Continue Reading

Learning the Lingo goes to CSCW

Jahna Otterbacher and I will be headed to CSCW in February to present our paper - Learning the Lingo? Gender, Prestige and Linguistic Adaptation in Review Communities. Our findings suggest that women do participate in online communities but that their contributions get buried and stay mostly invisible. See the Room for Debate feature Where are the Women in Wikipedia? for a popular press take on the problem. Read the abstract after the jump. Continue Reading