Social network analysis and innovation research

Social network analysis as a creative methodology of research and exploration was invented in 1960s and 1970s by social psychologists. It has been furtherly developed with multiple disciplines such as mathematics, statistics and computing. As a result, the collaboration led to a rapid development of the SNA. Currently, it has been used in many researches and disciplines. 

The concept of generating innovation by using SNA is simply make use of the fundamental principle of the SNA, which is embedded level/units researchers want to innovate into the SNA system and finally measure from different levels such as the node-level, the local-level, and the network-level to judge if the possible innovation could be realized. 

Data required 

The use of SNA in innovation research does not have specific requirement for the data because innovation is a thing that never been proved or conceptually existed. Whatever researchers want to innovate, they just need to collect, filter and organize the data. For different research areas, data requirements are different. Anything could be in the dataset including persons/individuals, teams, organizations, concepts, patents, etc. Data will be imported to the SNA system and analyze, finally SNA will export those measures. Researchers use those measures to determine if the innovation is possible to realize and achieve. Or if those measures proves that the innovation is valuable to realize. We can say the the acquisition of data for innovation research is easy. 

Network measures 

There are four most important measures for the use of SNA in innovation research which are network density, centrality, betweenness, and centralization. There are also four types of performance measures which are robustness, efficiency, effectiveness, and diversity. Those concepts grouped with several measures with various corresponding advantages and disadvantages regarding the use. 

Network density

This rate is expressed in a form of number from 0 to 1. Density is used to explain the number of ties in a network and the connectedness in the network. If the density is close to 1, the network is defined as dense. Oppositely, if the rate is close to 0, it is a sparse network. 

Centrality

Centrality includes two levels which are local and global. It is expressed as a rate from 0 to 1. Local centrality considers ties that directly connected to the node, however, global centrality centrality considers indirect ties that are not directly connected to that node. 

Local centrality measures are about how many nodes are connected to a particular node; Global centrality measures on the distance among different nodes. 

Betweenness

It is a further developed concept and measurement of centrality. Betweenness explains the extent that a particular node lies between other nodes in a network. A node in a network with few ties may play an critical role and be central to the network. 

Centralization

It explains the extent to which a whole network has a centralized structure. Centralization describes a fact that the extent to which the connectedness come from the density is organized around particular focal nodes. It is an very important complementary measures. 

Robustness

It is measured by studying how it become fragmented as an increasing fraction of nodes is changed or removed. 

Efficiency

Efficiency is a term that is used to determine if nodes can access other nodes instantly. It aims to reduce time and energy spent on contacts.

Effectiveness

Effectiveness is used to make those cluster of nodes reached through non-redundant contacts.  However, each cluster of nodes are independent. 

Diversity

Nodes must be diverse in nature if the performance wants to be good. 

SNA is used to explore causal mechanisms related to innovation research. It could help researchers diagnose problems that cannot be realized under a routine methodology. Researchers will have some surprising findings that cannot be derived by routine research. Measurements will lead researchers to a new area that never has been explored. New interviewees will soon be identified and researched. As a result, it is probable that innovations will be generated. Then management will make use of the innovation and make changes on the organization. 

Xufeng Hu-MIB