Polarised discussions feature two very dense divided groups and tend to centre on controversial subjects such as politics. The two groups do not actually directly argue, or converse with one another, despite talking about the same subject. Instead of actually arguing they end to ignore one another, using different links and hashtags.
Tight crowds are to a certain extent the opposite of the polarised crowd. While groups are dense; the links between the different groups are extremely strong. The number of so called ‘isolates’, people who have no connection to other people in these groups is extremely low. They also directly converse with one another regularly. Topics within this group are often based on mutual interests such as hobbies or sports.
Many customer complaints are now dealt with on twitter. To handle this a spoke and wheel pattern has emerged. The hub based around the customer service account communicates with many unconnected outward spokes. This group contains many ‘isolates ’ as the users are not connected to one another. However, the support network also contains small groups who will converse about the quality of the service they are receiving from the hub.
Broadcast networks usually focus on news from major providers or famous political pundits. The broadcast network is similar to the support network, in that it also has a hub and spoke structure, of largely unconnected individuals. But the big difference is that the users do not talk to the hub, instead they broadcast the information outwards, usually using retweets. Similar again to the support network, some small groups may form to discuss news, but members are mostly ‘isolates’.
Community clusters also focus on news and other popular topics, but develop a number of smaller hubs. Each of these hubs has its own audience, influencers and sources of information. Although there are numerous conversations between different groups of people, it also creates quite a number of ‘isolates’.
Brand clusters develop around large companies, products and celebrities. This group generates commentary from largely disconnected or ‘isolate’ groups of people about a brand. The larger the brand the more disconnected the users are likely to be. They will connect and comment on the brand, but do not tend to connect with one another.
In a previous article I discussed the power of Twitter and whether it was in fact too powerful. Even to the point where it can predict flu outbreaks by analysing tweets. This new article will examine the recent study which, suggests that tweeters can actually be divided into six different groups:
- Polarized Crowds
- Tight Crowds
- Support Network
- Broadcast Network
- Community Clusters
- Brand Clusters
The Study Explained
The study by Ben Shneiderman at the University of Maryland is the first of its kind ever carried out and took roughly four years to complete. By analysing tens of thousands of tweets from all around the world, on all different topics, the researchers were able to map the pattern of Twitter conversations into the six previously mentioned groups. The patterns were based on topic, information, conversation influencers and social network structures. According to Shneiderman “What we’ve done is to provide a visual map of the Twitterverse that will ultimately help others to better interpret the trends, topics and implications of these new communication technologies.”
With that in mind let’s take a closer look at the inner workings of the different groups. Which group do you think you fit into?