Computational Models of Social Capital

Friday, 24. February 2017

By Rajesh Sharma, The Queen’s University of Belfast, UK

 

In our previous post on social capital, we described and discussed the concept of social capital and how it can be useful for career development in workplace settings. In summary,

social capital is the capital (or value) of an individual due to its social contacts and resources of its social contacts.

The term resource is subjective and can refer to, for example, the domain expertise of a colleague, financial or political power, or even the number of followers on Twitter.

Social networks are an example of complex networks [1]. Importantly, for calculation of social capital, not only do we have to consider the edges (connections) between nodes (people), but must also take into account the type of edges and the types of resources that a node possesses. Thus, a natural question is: can we propose a computational model for the calculation of social capital? Intuitively, the answer is yes as computational models are needed for complex settings which further explores the power of computers [2].

We propose to integrate three well established theories that have been proposed in the past [3, 4, 5] into our proposed computational mode of social capital. These theories mainly take into account the network structure [3, 4] as well as the resources possessed by nodes [5]. Our idea is partially inspired by a seminal work from Seibert et. al. [6], where the researchers performed an empirical data analysis by exploring all three theories. Our model will exploit the resource theory of Lin [5]. In addition, our model will include concepts based on social network structures such as (i) weak and strong ties [3] and (ii) structural holes [4]. In other words, the position of the node in the social network can also play an important role in measuring its social capital.

1. The Concept of Weak and Strong ties: Social connections can be categorised as strong or weak [3]. The strength of a tie depends on various factors such as a combination of services, the amount of time, the intimacy, and the emotional intensity between two subjects. It has been shown that new information generally flows from the weak ties. In terms of social capital, this can be translated that if a node has strong ties with all of its connections, then all of its connections will be considered as of equal importance or will have expertise in the same domain. However, in cases where new expertise is needed, it is highly likely that it will come from a connection with a weak tie.

2. The Concept of Structural holes: Burt focuses on a particular network structure, where a node acts as a bridge between two disconnected components and thus, leading its importance in the network settings [4]. For example, consider the network in Figure 1. There are two tightly knit communities. First, consisting of nodes 1, 2, 3, and 4 and the second, consisting of 6, 7 and 8. Node 5, in the network acts as a broker between two communities or is a structure hole of the network and thus its (closeness) centrality value is very high. In simple terms, any information must pass through the node 5 if has to go through to the other community, which implies that the capital of the node 5 is high in the network.

Social Network
Figure 1: Social network with two communities and a “brocker”
 

3. Connection as a Resource: The social resource theory sees connections as a resource or pathway to attain a resource [5]. In his famous work, Lin argued that the connections (or network structure) alone are not sufficient for the calculation of social capital. However, it is important to verify if the connections do possess or can lead further to nodes which are in possession of the desired resource.

Our proposed computational model is based on the above three theories. For our initial investigations, we will take into account the social structure formed through workplace social contacts. However, this approach is not limited to workplace contacts and future work could include non-workplace contacts such as those formed with other organisations or through other social contacts (e.g. through social gatherings). Inclusion of contacts from non-workplace settings will naturally add complexity to the model as they represent another type of complex networks, thus forming a multilayer network [7]. For example, one of the networks could be workplace network and another layer could be a personal network (e.g. clubs, dinner), family/acquaintance networks, etc. The consideration of multilayer networks will make the model richer for calculating social capital.

References

[1] Newman M. E. J. (2003). The structure and function of complex networks, SIAM REVIEW, vol. 45, pp. 167–256.

[2] https://www.nibib.nih.gov/science-education/science-topics/computational-modeling

[3] Granovetter, M.S. (1973). The Strength of Weak Ties. Amer. J. of Sociology. 78 (6): 1360–80.

[4] Burt, Ronald S. (1995). Structural Holes: The Social Structure of Competition. Cambridge: Harvard University Press.

[5] Lin, N, (1999). Building a network theory of social capital, Connections, vol. 22, no. 1, pp. 28–51.

[6] Seibert, S. E. , Kraimer, M. L., and Liden, R. C. (2001). A Social Capital Theory of Career Success, The Academy of Management Journal, vol. 44, no. 2.

[7] Kivelä, M., Arenas, A., Barthelemy M., Gleeson, J. P., Moreno, Y., Porter, M. A (2014) Multilayer networks. Journal of Complex Networks, vol. 2, no. 3, pp. 203–271.