Matt Shapiro is presenting our paper – Going “Bald on Record”: Relationships Among Public Officials’ Social Media Behavior and Language Use – at this week’s Korean Association for Public Administration and American Society for Public Administration Joint International Conference. Can’t make it to Korea for the talk? Read the abstract after the jump. Jahna Otterbacher also contributed to this paper. Continue Reading
Wonder what those politicians are spending all that time on Twitter talking about? Well, here’s what they were talking about over the summer:
Data for this Wordle came from tweets between June and August from the 389 members of Congress whose Twitter accounts I could verify.
Many of Chicago’s Aldermen, and certainly the Mayor, have adopted Twitter. Using all the tweets they posted, and all those posted by others who mentioned them, I was able to make the following graph of Chicago’s politicians on Twitter.
In this image, orange discs are Mayor Emanuel’s two Twitter accounts, blue discs are Aldermen’s accounts, and gray discs are neither (I want to say constituents, but I can’t assume all are). The opacity and weight of the lines connecting them depend on how often the two people talk about or to each other on Twitter. The size of the discs is determined by their in-degree, or how often someone talks about them. We see that two Aldermen figure prominently in the social media conversation, and both respond to people pretty frequently (outgoing arrows). What this means yet, I’m not sure, but it may indicate something about the responsiveness of politicians. The next step is to determine whether any of the properties we can identify in the network relate to things in the real world, e.g., voting behavior, responsiveness, public approval.
Jahna Otterbacher, Matt Shaprio, and I just received funding from IIT’s Social Network Workshop to continue our study of politicians and social media use, so stay tuned.
It seems every time I embark on the qualitative data analysis (QDA) stage of a project, I get frustrated by QDA software. My approach to data analysis is iterative and collaborative. Existing QDA makes both of those approaches obscenely difficult (if possible at all). Hence my frustration.
I want to be able to edit transcripts even after I’ve loaded them into the software (e.g., if I notice an error in the transcription). I also need to be able to share my work with colleagues who use a variety of computers including PCs, Macs, iPads, and Android phones. Continue Reading
You can tell by today’s posts that I’m getting something started on EC2. You can also read what’s already installed on the free instance or how to get started. PHP and MySQL are not installed on the free instance, so this post is about getting setup so I can run 140dev’s excellent Twitter Database Server. Continue Reading
I’ve been having trouble finding out what’s installed and/or running on the free Amazon EC2 instance, so I’ll use this post to list the processes and installs as I uncover them.
Default versions of stuff I use:
- Python 2.6.6
- ruby 1.8.7 (2010-12-23 patchlevel 330) [i386-linux] (You’ll need to install Rubygems yourself)
I’m setting up some free Amazon EC2 instances so I can run my Twitter data collectors without violating the terms of my Dreamhost hosting agreement or worrying about losing power on my office machine (happened just last week). Despite the many Google results for “getting started with ec2 on a mac”, the process is not that complicated.
Matt Shapiro and I are collecting all tweets written by and mentioning members of Congress and Presidential appointees so we can study how they frame issues in social media. We’re especially interested in discussions related to climate change and comparing framing by members of Congress with appointees’ framing.
Out of the 961 public officials who could be tweeting, we found 471 Twitter usernames (actually, a few less than that because of duplicates like @gopconference). You can follow our lists on Twitter (via CaSMLab’s lists) and see data about who we’re following.Â Our list of appointees and members of Congress are current as of June 9,2011. We’ll continue collecting through the 2012 elections and hope to add candidates starting in January.
Thanks to Adam Green at 140dev.com, we have a very simple PHP database server collecting and parsing tweets by all the people we found. Adam also provides a barebones display of the collection we’ve made available on my lab’s website. We’ll be pretty-ing up that site in the future.
I’ve started following a bunch of TV actors, showrunners, and writers on Twitter. So many of those I follow have been tweeting about TwitChange.com that I got curious about the celebs’ Twitter networks and the potential reach the mentions and retweets people buy will have. With so many followers (most have >10,000), processing those networks takes a really long time, but I’ve dedicated a PC to gathering and analyzing networks for Amy Brenneman, Nathan Fillion, Michael Chiklis, and Kate Walsh. I picked two actors from Private Practice accidentally (well, conveniently) because I’m analyzing the Private Practice cast and crew networks for another project. Anyway, see the networks below (currently only Amy Brenneman is done, the rest are still processing). Continue Reading
I’m just getting my lab at IIT up and running. We are now the CaSM Lab (Collaboration and Social Media), and I invite you to learn more about us on our website. In just the first two months, I’ve been joined by two master’s students and one undergraduate. Many of our efforts will focus on Overherd this term, but we have other projects related to social media and collaboration in the works.