It’s like magic! But what does it mean?

UNU Africa Networks consolidated no lable1I’ve been mapping networks for some years now but in the past months I have used a more traditional approach to social network analysis for the first time. In a project with the United Nations University, we (Nancy White and I) are tasked to find out how their networks are structured and to initiate a debate around improving networks for increasing impact in Africa.

As the UNU is spread around the globe, we chose not to visually map the networks, but send out network questionnaires, basically asking questions like:

  • Who of the people on this list have you ever interacted with?
  • Who of these people do you currently collaborate with?
  • Who would you like to work with but have not approached yet? etc.

The questionnaire was administered by the people at Cross Analytics (at 4 US$ per participant they did a pretty good job in terms of support etc.) and all I had to do was wait for the results to be delivered to me in the square matrix format I need to enter it into Visualyzer.

I was absolutely excited when I got the data because it did look like magic – out of the blue these interesting networks emerged and I could play around with them, exploring the structure, observing how they evolved over time and guessing what it all meant.

And that is the biggest difference between participatory network mapping and a survey approach: With the survey you get the network structure but little of the background, the how and why. Sure, you can add non-network questions to the survey to elaborate the background. But often I only see the questions emerging while I draw the network, which means that after analyzing the survey results, I have more questions than answers… Confused but on a higher level… In this case that doesn’t hurt: The survey is planned as a first step in a longer process and an entry point for discussion, so I will get to ask all my questions.

But how do researchers do it who only rely on one course of surveys? How do they find out what the network structures mean? Why doesn’t this drive them crazy?

4 Responses

  1. Loved it. Every single word.

  2. It is not only magic, it actually works ! Some years ago, while leading a local government training program in the Balkans, we conducted a series of training-of-trainers workshops. At the start of each session, we’d put up a white board and ask TOT participants to indicate which other participants they had followed up with, what new contacts they had made, and where they hoped these interactions might lead in future. The result looked something like the inside of a golf ball ! It was fantastically illuminating for them and for me – and I’ve sworn by such similar exercises ever since.

  3. the map you created for the UN University project is beautiful and intriguing. I’m working on an article right now with Jessica Clarke from the Center for Social Media at American University discusing the impact of multi-platform projects my organization (AIR) generating

    I’d like permission to use the image, with attribution, in the piece. Please contact me directly at I would have sent email direct to you, but couldn’t find an address.

    congratulations on fascinating and excellent work.

    — ss

    Sue Schardt, Executive Director
    Association of Independents in Radio, Inc (AIR)

  4. Hi Eva,
    the map you made for the UNU really interests me. I am working at UNU-CRIS currently and in order to do my research for my thesis I am learning how to work with web crawling to define unknown stakeholders.

    Can you give me some more information on the research you did?



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: