Some stuff that I encountered in the past weeks. Somehow, it’s all networks-related.
- In the latest issue of AJS, Andreas Wimmer and Kevin Lewis use the ‘Tastes, Times, Time’ Facebook data and ERG-models to disentangle the various network formation mechanisms that generate aggregate racial homophily.
- While the above article is based on a very specific subset of the Facebook network, a recent working paper by Minas Gjoka, Maciej Kurant, Carter Butts, and Athina Markopoulou evaluate sampling methods to makes inferences about the entire Facebook network. Interestingly, they managed to obtain a uniform sample of nodes to use as a benchmark. Not only do they come up with a sampling method that produces unbiased results, but almost as an afterthought, also produce some descriptives of the Facebook network. Among other things, they show that the degree distribution of this network does not follow a power law. In my opinion, these results alone would be enough for a publication in a major journal.
- The open-access game theory journal Games has a special issue on ‘Social Networks and Network Formation,’ including pieces on public goods, coordination, and learning in networks, among other interesting stuff.
- James Rauch reviews Matthew Jackson’s ‘Social and Economic Networks’ in the latest issue of JEL, and is, perhaps surprisingly, rather skeptic about network theory as presented by Jackson. Quite refreshing, even if I do not subscribe to his critique.