Thesis: 4.6 Contribution to the literature

An paper with Bernie Hogan based on this chapter is available. It was written after this chapter and contains a number of changes.

The contribution of this project is two-fold. First, we proposed a general method for analysing the evolution of longitudinal network data. The method is based on a regression framework that enables us to test how different mechanisms jointly drive network evolution. It is not the first time this regression framework has been used to study network evolution. Powell et al. (2005) studied tie generation in an inter-organisational network of pharmaceutical firms. However, due to the nature of their data, the model they applied was not generalisable to other datasets. More specifically, they combined one- and two-mode data. Moreover, they included four different terms to account for preferential attachment, and calculated homophily based on similarity with previous partners. In addition, the data were only recorded at yearly intervals and, therefore, the dependence among ties occurring in the same year was not taken into consideration. The method that we proposed in this Chapter is more general and a special case of the actor-perspective applied by the SIENA model (Snijders, 2001).

Second, we have empirically tested six growth mechanisms independently and jointly in binary and weighted networks constructed from a self-organising virtual community. These networks are over four times larger than the one used by Powell et al. (2005) and the exact sequence of ties is known. Two often overlooked aspects of network analysis are reciprocity and reinforcement. Since communication in the community was directed, and repeated interactions were recorded, we could test for these two aspects. Our findings allow for a better understanding of the underpinning principles that guide interpersonal dynamics in an online environment. In particular, we found little or no effect of triadic closure depending on model specification. This is surprising as the number of common friends has generally a strong effect on the evolution of social networks (Steglich et al., 2007). Nevertheless, this result comes from an online social network. Online behaviour might differ from offline behaviour, and different mechanisms might underpin tie formation in the two contexts. By shedding light on these principles and results, managers and moderators of online communities are able to device better strategies for improving communication and knowledge transfer.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Subscribe to the comments via RSS Feed


%d bloggers like this: