A Longitudinal Social Network Analysis of German Politicians' Twitter Accounts

Politicians running for an office in the German Bundestag election in
2009 for the first time massively have taken the liberty of getting
involved with social networking and becoming their own "reputational
entrepreneurs" in social media (Fine, 1996). While reputation in
social networks does not emerge from good work directly so much as
stories about the work, the key to building reputation is to get
people in closed networks talking about oneself (Burt, 1999, Gladwell
2000). The created reputation and its stability in such a network is
clearly not independent of network closure (Burt, 2000) mechanisms,
which reduce the risk of trust among people and carry on reputation
from one year to the next. In order investigate those mechanisms in
context with the German Bundestag election, we took daily snapshots of the social network formed from over 650 politicians on twitter in a timeframe of 4 weeks. By tracking almost 10.000 connections of over 650 twitter accounts and monitoring over 240.000 tweets we were able to investigate how and with whom politicians established connections and which topics they discussed. Using statistical social network analysis methods (Snijders et Al., 2007, Carley et Al. 2009) our results show that there is indication for closure among members of their own party. They majority of connections are established between members of the same party while connections between different parties are significantly less represented. The analysis of the exchanged tweets shows that conversation is directed towards members of the same party and mentions of opposite parties can often be found.