The most recent edition of First Monday takes a look at some online social networks. One researcher surveyed a thousand Facebook users at the University of South Carolina and found, “Essentially, Facebook appears to operate primarily as a tool for the facilitation of gossip.” Really? I hate to make fun of academics but please tell me no precious grant money was used to come up with that finding!
There's another piece that does some groovy statistical mapping of users in some other networks. Here are the conclusions of that researcher:
- Membership growth: As a rule, the membership of networks will grow at a linear rate. The rate of growth can be affected positively or negatively by publicity. The only exception we found was Hipstir, where the network was in decline.
- Profile deletion: Members rarely delete their profiles from social networks. When there is a clear and easy mechanism for deletion, some people will take advantage of it, but they represent a tiny fraction of the population of the network.
- Relationship dynamics: Users add relationships frequently, and in most of the networks we looked at, the networks grew denser, with relationships growing more quickly than the number of new members. Users will also delete relationships, but at rates that are orders of magnitude less frequent than they add relationships.
- Social disconnection: The percentage of users who are disconnected from the main cluster, or who are completely socially isolated, varies widely among social networks. If non–social features of the site are more important, the percentage of socially disconnected users increases.
- Centrality: Users who tend to be toward the center of the cluster, not surprisingly, are the users with the most number of friends. They also tend to be users who have been active longer in the network.