Monday, April 15, 2013

Social network analysis: how to use it for online networks using social media?

One of my huge interests is networks and how they can thrive online. Before the internet existed I was working in Mali with farmer organisations. We used to do Venn diagrams (by cutting paper or sometimes drawing in the sand) to find out how they collaborate with other institutions. It was always fun to do. The most interesting part was where the farmers discussed how big or small (= importance) and organisation should be. Very often the agricultural extension service was put us as a small circle because they were not very useful for the farmer organisations. By doing this, you could gain a lot of understanding about the collaborations and where improvements might be needed. For instance, we worked on sericulture in one district and found out that major institutions like FAO were not involved in sericulture development and hence organised a national meeting to engage national institutions.

BUT now.... we are all online. I still work regularly with networks and communities. Social Network Analysis seems to be the Venn diagramming of the online world. I know Social Network Analysis (SNA) is a whole field producing all these impressive diagrams, but I've never done an SNA because somehow it seems complex and time-consuming. Somehow these graphs seems to have their attraction because they look scientific, however I think back to how the farmers discussed where to put a circle and wonder how 'right' the answers are.

To learn more about SNA as a tool I decided to do the Organisational Network Analysis course by Patti Anklam. Unfortunately this is an extremely boring e-learning course of the traditional 'listen and do the test' type. There is no interaction at all with the teacher or other participants. So we set up our own little group of 4 to discuss after each module. Tomorrow we'll talk and agreed to start writing some thoughts in a blogpost. So far the course explains how you can use SNA in organisations, how to design an SNA project, the network patterns and metrics and and tools for SNA (UCINET and netdraw are discussed). If you want to read Josien's blog on the same topic you can find it here.

Questions I have after the first 3 modules;
  • Is it possible to do network analysis without collecting data by using social media connections? The course focusses on organisations and assumes you can invite people to respond to an online survey (or paper survey). Would it also be possible to analyze online social networks by for instance looking at Twitter links? In the book Netsmart by Howard Rheingold offers the example of analysis of people who participate in a Tweetchat using NodeXL software.What are examples of this kind of analysis (without collecting additional data).
  • How much is the time invested in doing the SNA and does it justify the results in terms of surprises and new insights that would not have been possible to gain in other ways? After the first 3 modules I get the impression that if you know very well how to work with the software it should not be too time-consuming but learning the software will take some time.
  • Would it be possible to make the process more participatory? The farmers used to have extensive discussions to visualise their reality in the Venn diagram. Would it be possible to do a SNA where you invite people to make their own models before presenting the diagrams and metrics that look so scientific that people may not dare to challenge them?
By the way, if you are interested in SNA - there seems to be an active community around the #sna hashtag on Twitter. 


mirjam said...

You would love the SNA course from Coursera. Not easy, quite some math, but lots of fascinating stuff to learn.

Joitske said...

Thanks for the tip! I will definitely look at it after graduating here :)

Nancy White said...

For the use of social networks as the collection point for data, look at NodeXL and the work of Marc Smith

Also, of course you know Eva Schiffers NetMap Toolbox for a participatory approach, right? said...

I'm particularly interested in the socio-cultural analysis but have dabbled in the more mathematical side on a project but am by no means a mathematician......I was hoping to find said folk here for possible future collaboration.

In th emeantime thanks for starting the group.

Joitske said...

Do you know the networkweavers group on facebook? you might join that group!