Is political discussion possible in online chatrooms? It absolutely is, argues Jennifer Stromer-Galley in a recent article. She and Anna M. Martinson analysed chatroom conversations to see whether or not the particularities of this online form of communication hinder meaningful, coherent and engaged discussion about politics.
(Previous research tend to claim that chat is not really suitable for such discussion, as messages are short and fragmented, arguments are underdeveloped, and random, wild changes in topic are rife.)
Without going into details of methodology, the idea was to compare chat exchanges from various topics of Yahoo! chat, and analyse their coherence, taking individual chat messages as units of analysis. The topics involved were politics, car racing, entertainment in general, and support for cancer patients.
Perhaps surprisingly, political discussion was measured to be the most on-topic by a large margin (93.5% of the messages were coded as on-topic), and it was the political discussion too that proved to be the most coherent (79% of messages relating to each other in such a meaningful, constructive manner that serves the development of conversation).
|Discussion topic||Auto racing||Cancer support||Entertainment||Politics|
|Num. of participants||40||39||22||26|
|N: first 500 lines of chat entered during a 2-hour recorded period on Yahoo! Chat|
This suggest that
users can sustain relatively coherent interaction on politcal talk [in chatrooms], suggesting [that] chat technology may not be an inherently problematic medium for political discourse.
The research also underlines that the topic of conversation affects the quality of conversation.
Two brief notes:
- If you consider the number of people having participated in the various chat sessions, it can be seen that political talk did not come to be more coherent than the others because there were fewer people involved in it. However, what the research doesn’t tell us is the distribution of messages among participants – i.e. it might be that the political conversation involved fewer people who actively participated in the conversation – and not just turned up to say “hi.”
- Coherence was measured on 2 levels. First, on the level of “organizing topic,” that is, looking into what percent of messages was on- and off-topic; second, on the level of “interactional topic,” looking into how the messages related to each other in developing the conversation. Not that these two do not depend on each other. An offtopic message can generate a perfectly coherent discussion (which has nothing to do with the organizing topic of the chatroom.
Jennifer Stromer-Galley, Anna M. Martinson (2009): Coherence in political computer-mediated communication: analysing topic relevance and drift in chat. Discourse & Communication 3(2), 195-216.