Let’s continue the discussion using the same network as in the last part, shown in the figure above. If you recall, the first model for gaining social capital in an organization suggested that a person with a high clustering coefficient (such as the red circle on the left) would benefit from having a highly connected neighborhood.
The alternative model, propounded by Prof. Burt, looks at how information flows through a network. It is well known in sociology and social psychology that information flow within a connected group is faster than between groups, mainly because the connections are typically denser within than between groups. In the figure above, if we consider the blue and red clusters to be different groups, the above statement would seem to most people as stating the obvious!
Burt calls the lack of connections between groups as “holes” in the social structure or structural holes. Thus in such a situation, the person who is able to control the information flowing in between groups, across these holes, gets a competitive advantage. As he says “the structural hole between two groups does not mean that people in the groups are unaware of each other. It only means that people are focused on their activities to such an extent that they do not attend to the activities of the people in the other group”. Structural holes are thus an opportunity to broker the flow of information between the two groups.
Thus anyone who is in a position to regulate this flow of information between the groups stands to gain immense social capital. In the network given above, the large yellow circle and diamond are the ones who control this flow of information between their respective groups. Hence, as we have seen in the first part, the betweenness centrality scores (BC) would be extremely high for these individuals.
What does this mean for organizations? Does data really back up this contention that people who regulate the flow of information tend to do better? Do they get paid better? Do they get higher bonuses? The better assignments? We shall look at the data in the next post.