Note: This post represents the synthesis of the thoughts, procedures and experiences of others as represented in the articles read in advance (see previous posts) and the discussion among the students and instructor during the Advanced Analytic Techniques class at Mercyhurst University in November 2015 regarding Social Network Analysis as an Analytic Technique specifically. This technique was evaluated based on its overall validity, simplicity, flexibility and its ability to effectively use unstructured data.
Social Network Analysis (SNA) is a methodology in that the mathematical assessment of a network will produce an estimate, typically involving which are the most important or influential nodes. This technique involves the examination and assessment of a network, which is composed of individuals (nodes) and their relationships (ties). Utilizing this type of analysis, a visualization of the network can be developed. SNA can be used to determine the robustness and efficiency of a network, and can therefore be used to move information through the network or destroy the network all together.
- Can provide a holistic picture of a network
- Maps the spread of ideas and can determine who is the most influential individual(s) in a network
- Provides an alternative view where attributes of individuals are less important than their relationship ties
- It is highly quantitative and mathematically valid
- SNA in a matrix form can provide a quantitative score of relationships, although it will have bias associated with it.
- Identifies the type of network they are analyzing: centralized or decentralized
- Can help analysts identify the most powerful nodes in a network
- Can be used to simulate how information ripples through a network
- Can be used to identify intelligence gaps
- This tool can only be used in some situations, not all cases are applicable for social network analysis
- Defining relationships and understanding what that relationship means is difficult
- Visualization does not always tell a clear story therefore making it difficult to communicate to a DM
- Only allows for one relationship between two people to be defined, so it must be defined carefully
- SNA for ambiguous organizations may cause confirmation bias
- Constructing a matrix can be time consuming
- Matrix attributes can be subjective and often biased
- Step 1: Identify a pattern of relationships or lines of communication you want to analyze. In other words, identify a possible network for further exploration.
- Step 2: Input all members of this group/network/etc into a matrix. On the left side will have all members who initiate communication and on the top will be all receivers of communication.
- Step 3: Score the relationships on whatever scale you deem fit. Example: (Joe talks to Maria, that communication for Joe is a score of 5. His communication to his enemy may be a -5.) The highest scores of all the rows would be members that would be key nodes and members would be considered the main communicators. The highest scores at the bottom are the members who receive the most information/communication.
- Step 3: Input the data (matrix) into a software program to generate the network map. On a small-scale individuals can draw out the map based on the matrix or via known dialogue/intelligence.
- Step 4: Based on the matrix and the visual network map, identify the key relationships and the strongest communication nodes.
Personal Application of Technique:
Practical Exercise 1
For this exercise, we used four people to establish a small network flow focused on the flow of information. The four people were: Professor Wheaton, Andrew, Dan, and Katie.
1. Design a map based on the following communications:
- Prof. Wheaton: Andrew, go tell Dan & Katie to meet tonight for a super secret meeting.
- Andrew: Katie, make sure Dan gets to the meeting.
- Andrew: Dan, you & Katie better be at the super secret meeting.
- Prof. Wheaton: Katie, did Andrew tell you about the super secret meeting?
- Dan: Hey Katie, we have a super secret meeting to go to.
2. Make a matrix
Prof. Wheaton: project leader, not a micro-manager
Andrew: asst. project leader, more of a micro-manager
Dan: team member, friends with Andrew
Katie: team member
Practical Exercise 2: Family SNA
1. Imagine big news in your family and how news would spread
2. Pick 4 to 5 people to assess the network
3. Draw the network map
4. Draw the matrix
5. Fill it in & scoring it based on your knowledge of personal relationships