What is a Social Network Analysis? Social network analysis (SNA) can be defined as a, “distinctive set of methods used for mapping, measuring and analyzing the social relationships between people, groups and organizations,” (Blanchet & James, 2011). SNA can produce maps like the one featured below, and provide statistical measures of relationships between actors. In SNA maps, the nodes represent the different actors in the network, and the lines represent the relationships between the various actors. The size of the node often represents the relative importance of that actor in the network, and the thickness of the connecting line denotes the strength of the relationship.
www.partnertool.net
Why conduct a SNA? SNA is a unique way of analyzing relationships and their implications for practice that incorporates both qualitative and quantitative data. SNA can be used as a research or evaluation method to understand how different actors are working together, sharing resources, or communicating across a network.
What kinds of research questions can a SNA answer? SNA can answer several types of research questions including: •Are the right connections between actors in place? Are any key connections missing? •Who are playing leadership roles in the community? Who is not, but should be? •Does the collaborative measure well on indicators predictive of successful networks? (e.g. high levels of trust, mission congruence, resource contribution, transparency)
How is SNA conducted? Conducting a SNA usually involves surveying actors about their relationships with other actors in the network and then mapping these results. There are several considerations to take into account when designing a SNA. To help guide these decisions, please see the flow chart on the next page.
Social Network Analysis Design What types of actors will you survey? The units or actors in SNA can be individuals, professional groups, departments, organizations or larger system level units.
What types of relationships do you want to analyze? SNA can measure many types of relationships between actors, including communication flows, resources exchange or social support.
What do you want to know about the actor relationships and the network? There are many SNA indicators that can be used to qualify relationships between actors including strength, reciprocity, centrality and betweenness. There are also indicators used to measure the properties of the network, such as size, density, clusters and stability.
What type of analysis should be used? There are two main types of SNA:
Whole Network Analysis Whole network analysis aims to understand the connections between all actors in a network. This type of analysis is done when the network boundaries are known and when it is possible to collect a list of all actors in the network.
Ego-Centric Analysis Ego-centric analysis focuses on identifying the complete network of a single actor or node. It can also be used to analyze a complete network when the network boundaries and membership are not known a priori.
Methodology: In a whole network analysis, a list of actors is surveyed about their relationships with all other actors in the network. It is important to get a near perfect response rate so that you can understand the degree of relationship reciprocity.
Methodology: Ego-centric analysis involves asking actors to identify actors with whom they are connected, either by listing from memory, or selecting from a pre-defined list (with room to add additional actors that are not listed). The SNA will then show how these individual networks overlap. A two-mode or bipartite ego-centric analysis measures how actors are connected to one another through an intermediary (such as an event or mode of communication).
Tools: PARTNER is a tool designed to conduct whole network analyses about public health collaboratives. It includes a validated survey that measures connectivity indicators and an analysis software that is very user friendly. You can also design your own whole network analysis survey through any survey platform such as Survey Monkey and then use an open-source SNA analysis software such as UCINet, NetDraw, Pajek, GEPHI or NodeXL.
Tools: You can design your own ego-centric network analysis survey through any survey platform such as Survey Monkey and then use an open-source SNA analysis software such as UCINet, NetDraw, Pajek, GEPHI or NodeXL.
