Playing with numbers…

Paolo asks for an example to clarify the post below. Let’s see if I can do that. The network picture is an information network, drawn by the actor “parboiler”, located in the middle of the picture. This is one of a number of interviews we want to look at.

The table shows the degree centrality values for this network. As you can see, the interview partner has an extremely high centrality value, because she knows most about her own links. Because of that it doesn’t make sense to compare her with the rest of the list. What does make sense though is to look at the rest of the actors she talks about, seeing e.g. that consumers have a higher degree centrality than their family (what ever that means in this specific case). Now if you are asking 20 par-boilers, you will find that (most likely) most will put themselves in a central position of their own information network, but still, some will just have a few information sources, while others have many more. Maybe you are interested in their absolute numbers of information sources, then you will compare the in-degree, out-degree or degree centrality (combination of in- and out-degree).

But if you want to know how connected they are as compared to the connection potential within their specific network, you look at the normalized centralities (last three collumns). They basically show: What percentage of the possible links (being linked to everyone in the network) does an actor have. The fact that the normalized degree centrality for parboiler is higher than 100% looks strange but stems from the fact that they add the normalized in-degree and the normalized out-degree. If you have collected non-network data (about the income, education, age or family situation of the women for example) you could then develop some ideas about why some women are better networked than others. Or what the effects of being well connected are. Maybe women with higher education seem to have more information links. Or women who don’t have husbands (because the ones with husband rely on the husband for most external information).

Obviously, the fact that two things appear together doesn’t tell you, whether one causes the other. Just like the old story of the ambulances at the site of an accident. Someone observed that the more ambulances you can count at the site of an accident, the more people died in the accident. So the logical solution was: Don’t send these ambulances and no one will die.

Everyone thinks he’s the center of the universe

And they don’t know that it’s really me…

That’s what you get, when you ask people to draw their networks, they (most of us) will draw a network that makes them look much better connected than most other people in the network. That does not necessarily show their delusion of grandeur but just reflects the fact that you know much more about your own links than about other people’s links.

We are running into this problem at the moment when interviewing female rice processors in northern Nigeria about their business networks: If you naively looked at these drawn networks, without knowing, which network actor was the interview partner, you would think that these women are the best connected people in the whole of northern Nigeria. So, if we still want to look at the quantitative measures, what do we do?

If we want to compare centrality values within the networks (Who is most central in this specific interview partner’s network?), we compare everyone’s but the interview partner’s value. Not that we would cut them out of the network when calculating (because then we would have an unrealistically disconnected scatter of actors) but we don’t include their extremely high centrality values in our comparison. If we want to compare between different interviews however, it might be very interesting to compare: How high are the centrality values of different interview partners and to ask e.g.: Is there a connection between how connected an interview partner sees herself and how successful she is in her business? For comparing between interviews it makes sense to use the “normalized” centrality values. They basically indicate what percentage of the highest possible centrality the actor has. So a normalized degree centrality of 70% means that the actor is connected to 70% of the actors in this network.


Do you know the feeling: You only realize something, after you actually said it? Today I was asked to write a few sentences about myself, as kind of an informal introduction. This is what I wrote:

“I’m German and live in Washington DC for 2 1/2 year now, before I lived and worked in Ghana, West Africa. At the moment I divide most of my time (and passion) between being a free-lancing facilitator (with focus on developing countries) and mothering my daughter Sarah, 10 months old. In the time that’s left between these two, I dance (tango), read (fiction), write (blogs), cook (vegetarian) and talk with strangers.”

What struck me when I read through it again is the “talk with strangers” part of it. Because it’s true, I do talk with strangers. A lot. Every day, if I have my way. My husband knows that a day when I haven’t left the house, is a bad day. And that is, partly, because I haven’t gotten my daily dose of strangers. Why do I like talking to them, with them? Because they are strange.

They tell me things I don’t know. For example: “Poodles are good dogs for people with allergies”. Random stuff. They inspire me. For example by talking about Personal Kanban (even though after talking with Jim Benson I can hardly call him a stranger anymore. But… come to think of it, that’s true for all strangers after you get to know them, see lyrics below). They have networks that I have not and might have answers to questions I didn’t even know I had (see above, in re. poodles…).

When I first talked with Melinda Blau about her book “Consequential Strangers”, I was intrigued, because I always love meeting people who are going into the depth of exploring something where I just had an inkling… It’s a very readable book about all kinds of reasons why people benefit not only from the close interaction with their loved ones but also and especially from the weak links, the acquaintances, the friends of a friend’s brother-in-law and the ordinary street-stranger…

And here is the sound-track to this post: Tom Waits and Bette Midler never talk to strangers…

Count ’em

One new year’s resolution:

Every day I will make a list of three things I am grateful for. Counting my blessings. Grand and mundane. My beautiful baby and the crunchy fennel and anything in between.

Maybe I’ll start yoga or meditating or running next year. For this year, that’s about as much time as I’m ready to set aside for striving for enlightenment. A three point list. You’ll see if I start glowing like an old fashioned light bulb and walking a few centimeters above the ground by the end of the year.

The truest truth?

In an African country (that I won’t name now, as it could be any and we are in the middle of the process) we are mapping people’s perceptions of who influences specific policy making processes. After a first look at the resulting maps we realize that there are two quite different narratives:

  • Some people say: The policy process basically happens in the president’s head. Maybe influenced by a few informal advisers. The activities of others who claim a formal or informal role in the process are basically just noise but don’t change anything.
  • Others say: There are a lot of different actors influencing the policy process. The president plays a crucial role but is only one of many actors.

The automatic researcher’s reflex is to ask: So who is right, whose truth is true?

But as this project is also about advising our partners on how to have a stronger impact on policy making in this country, this question might be slightly besides the point. For strategic planning it’s much more important to ask: So who believes what (including yourself)? Because if we know that, it will be much easier to predict people’s behavior and to know how to approach them.

Those people who believe in their own ability to have an impact, will be more easily drawn into a coalition to advocate for change than those who think that decisions are completely removed from their sphere of influence. Basically there are two different approaches if you want to get these two groups to take action:

  • For those who believe that it’s all in the president’s head: Focus on process. They will only act if they believe that they can have an impact. No need to overwhelm them with content (such as: we should change the policy to become this and that) before they believe that they can achieve change.
  • For those who believe that other actors can play a role: Get them on board to share your goals. Got get there, you can bring in all the content you want to share. But keep the process in mind and while convincing them to join you, shape the format in which you advocate for shared goals.

But… before you even start, the question is: Which narrative do you believe in? If you’re like me, you’ll have good days (we can do it!) and bad days (it’s all in someone else’s hand). But the fact that you took the job of impacting on this country’s policy let’s me guess that you tend to believe that there is at least a sliver of a chance. And even if you strongly doubt it at times, just go for some pragmatic optimism because if there is the chance that you have an impact (which you aren’t sure of) you will only realize it if you believe that you can.