In the past weeks I have been traveling and teaching and learning more than reflecting and writing. Now back home in Washington DC, I am unpacking my bags and letting all this experience in Bolivia, Kenya, India and Ethiopia settle, to see what the neat and nice bloggable lessons are that I can draw from this.
Especially when teaching Net-Map to consultants in Bolivia (for the Inter-American Development Bank) and giving a general Social Network Analysis course to researchers of the International Livestock Research Institute in Nairobi, Kenya, we had a lot of discussions about: What do you do with your results?
There are rather qualitative and rather quantitative ways of using network data and one thing that I observe in my reading is that a lot of the research oriented social network analysis has a strong leaning towards the quantitative approaches of calculating centralities, network properties etc. Which, I admit, is one of the coolest things about SNA.
However, in a lot of the ad-hoc, pragmatic and barefoot applications of SNA tools in the field of development, most of the conditions for a clean quantitative analysis are not given, you deal with incomplete samples, actors of different kind (individuals, organisations, functional groups) and multiplex formal and informal relationships that you want to understand in the specific social and cultural context. So purely quantitative analysis might be too much (methodological overkill) and too little (reducing the depth of understanding to pure figures) at the same time.
What if you would use your network map as an illustration for understanding and telling a coherent story? If you look at this network of information flow in the Ghanaian agricultural system for example; the question was: How does the information about a suspected outbreak of avian influenza move from the place where the chicken dies to the national level authorities?
The story I want to tell with this map is the following: If there is an outbreak on a small farm (black dot on the bottom of the right cluster), there is a lot of communication on the local level. However, just one actor (animal health technician) carries all the burden of bridging the gap between the local level and higher levels (district, region, national). So the red arrow indicates a potential break point of the communication chain and warns everyone involved to support this link as good as they can.
And because I want to tell this specific story, give this specific warning, I have not chosen the (quantitatively generated) spring embedding layout, that puts actors with most links in the middle, but have manually dragged the actor-dots around a bit, to make the precarious bridge more obvious.
Filed under: case studies