Social Network Analysis Overview

Social Network is a representation of the relationship between various actors in a network. Actors could be people, groups, organizations, computers, Web sites or any other information/knowledge processing entities. Actors represent nodes and relationships or interactions are represented by edges. Linkage between the nodes acts as channels for exchange of flow of resources.
Social network analysis (SNA) is the mapping and measuring of relationships, connection power and flows between actors leading us to identify how we can best interact to share knowledge. Data for SNA can be collected through surveys, email logs, discussion forums, phone logs, twitter etc.
In analyzing networks we evaluate the location of actors in the network and these measures provide insight into the types of roles and groupings in a network e.g. who are the knowledge holders or gatekeepers, connectors, leaders, bridges, isolates, where are the clusters and who is in them, who is in the core of the network, and who is on the periphery, with consequences for how the network ‘functions’ and the individuals within it behave.
Numerous SNA metrics are available for analyzing groups and the individuals within groups.

Metrics (measures) in social network analysis

Diameter: Diameter is the longest of all the shortest paths in the graph. If everything is very tightly connected, then the diameter will be very short.
Density: Density measures the potential of the entire network to talk to each other. It is the ratio of number of actual connections to total connections.
Degree: Degree of a graph is the average number of edges associated with a vertex that is the number of direct connections a node has.
Centrality
Centrality ,measures the importance of the vertex. If the diameter is long, and the average length of all the shortest paths that go through a particular vertex is short, then it’s not part of the diameter of the paths that define the diameter of the graph. And so, in some sense, it’s maybe less central or less important.
Degree Centrality: Degree centrality is a simple measure that indicates the overall number of connections for each actor in a network. It is calculated as the degree of a vertex divided by total number of edges. It measures the network activity of a node. Important nodes are those with high degree. However, more degree centrality is not always better. If a node acts as a hub and has connections only to those in its immediate cluster then that is not really a good thing. That node is not adding any value to the network in the real sense. What is more important is how do they connect to otherwise unconnected nodes. This brings us to the concept of betweenness centrality.
Betweeness Centrality: Betweenness centrality is a measure which indicates the ease of connection with anybody else in the network, in particular, to try to connect potentially small subcommunities or subclusters of the nodes that may be happening in the network. Node with high Betweenness centrality is called network broker. A node with high betweenness plays a powerful role in the network, it has a great influence over what flows — and does not — in the network. It can also act as a point of failure and a bottleneck for information flow.
Closeness Centrality: Closeness centrality is used to measure the ease of access to everybody else in the network. Closeness centrality indicates how quickly they can get to anybody else in a network. They have the shortest paths to all others — they are close to everyone else. They are in an excellent position to monitor the information flow in the network — they have the best visibility into what is happening in the network. Such nodes can act as leaders and are characterized by high achievement levels.
Network Centralization
A network dominated by a single or a few nodes is a very centralized network. Such networks are not desirable as they are prone to failure in event of damage or removal of the central node. If these nodes are removed or damaged, the network quickly fragments into unconnected sub-networks. A highly central node can become a single point of failure.
Peripheral Players
Nodes on the outer end of the network which are only loosely connected to the network play a key role of providing important resources for fresh information not available inside the network. Though they appear peripheral in context of one network they might have their own network and can prove to be a source of new ideas  and information from other network.

Benefits

Social Network analysis has shown that our position in a network and our level of connectivity often determines our opportunities, level of influence, creativity etc. The network can also be used to measure social capital – the value that is generated by contribution of the nodes in network.
Sone of the benefits which can be derived out of the social network analysis are as below:
Information exchange and sharing:
Tie strength is important in assessing the overall connectedness of actors in an environment and the likelihood that information will flow from one actor to another. Strong ties have long been considered to be conducive to the exchange of information. Those who are closely tied to others have more intimate ties and are more motivated to provide information to others. Information dissemination is fast and assured in such networks.
Innovation and opportunities
While strong and close ties motivates information exchange, it is better for individual success to have connections to a variety of networks rather than many connections within a single network to get wider access to information.
Mark Granovetter found in one study that more numerous weak ties can be important in seeking information and innovation. He argued that “those to whom we are weakly tied are more likely to move in circles different from our own and will thus have access to information different from that which we receive.”
Cliques have a tendency to have more homogeneous opinions as they share many common traits resulting in members of the clusters to be attracted. However, the clusters will lack any new knowledge if they are not connected to other networks or subcommunities. Network brokers or nodes connecting subcommunities are highly associated with achievement and creativity as they have access to fresh ideas.
Social Capital
An individual dominated network is not desirable as it leaves little room for knowledge creation or generation of social capital. Rather, a desirable condition is to have all peers in the network with a similar network position. In these situations we can see that each individual in the network is equally contributing to the network towards knowledge generation. At the same time, each individual is also integrated into the network.
Sense of community
The sense of community indicates the extent to which individuals feel to belong to a particular community they are a part of. Sense of community can be an important predictor of employee retention in organization setup and of student’s retention in universities.
Peripheral nodes that appear to be isolated or sparsely connected to the rest of the network have a lower sense of community with respect to the network they are peripheral to.
Shane Dawson has demonstrated that closeness and degree centrality are positively correlated with the sense of community while betweenness centrality is negatively associated. He argues that the students who are brokering between different small sub communities that are created in a network are typically investing much effort and time to broker between all these different connections. At the same time they don’t feel that they receive much in return thereby resulting in poor sense of community.

Achievement and Performance
It has been shown in many different studies that performance in general is typically associated with different network centrality measures. Those individuals who have better and broader weak ties are better placed to have better performance, find a better job, or academically perform better.
Outcome of a study showed that there was a strong significant association between closeness centrality, and eccentricity with the academic performance. While degree and betweenness centrality were not significantly associated with the academic achievement. This basically means that those students who are central in crossclass networks, that is to say those students who need to go through the lowest number of the other nodes to access some other node in the network, were best placed and had best academic performance.

High Degree More connections Activity or gregariousness Fast Information dissemination.Strong sense of community
High Betweeness centrality The fraction of all the shortest paths that pass through it. Network Broker Great influence over what flows in the network.Access to wide range of information.

Innovation and opportunities

High Closeness Centrality Average distance of all its shortest paths. Leaders, Network monitor Higher sense of community.Achievement, opportunities
Centralized Network Single or a few central nodes Fragile network Prone to failure. If the central node is removed

Applications

In the field of medical research, social network analysis can be used understand the pattern of spread of epidemics and measures to control them.
In the socio political environment, it can provide insight into the interplay between communication, sharing of information, spreading of rumor thereby allowing for preempt measures to prevent rioting, spreading of communal disharmony etc.
In an organization it can help in decision making process by enabling the managers to understand the pattern of knowledge sharing. Identification of who knows who and who might know what can lead to better team building. Identification of teams and individuals playing central roles or thought leaders can boost innovation. Similarly, knowledge bottlenecks can be identified and strategies be worked out to improve knowledge flows. Employee retention can be improved by gauging their sense of community through social network analysis.
In areas of marketing, rate of return can be improved by identifying the appropriate locations in a network which can enable quick dissemination of information about a product and influence other actors to adopt them.
In the field of learning, instructors can find out about level of interest about a particular topic among the students and can accordingly strategize to improve the course content. Learning analytics and social network analysis can help instructors see what is happening inside of their classes. If the learners are having a low level of sense of community instructors can intervene and facilitate interaction among isolated students.
Social network analysis can be used to monitor creative capacity of students and develop it. Social networks play a key role in hiring, in business success, and in job performance. Networks provide ways for companies to gather information, deter competition, and collude in setting prices or policies.

Reference:
http://www.orgnet.com/sna.html
http://www.kstoolkit.org/Social+Network+Analysis
https://weadapt.org/knowledge-base/adaptation-decision-making/social-network-analysis
http://www.leydesdorff.net/betweenness/

Advertisements

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