When doing Net-Maps with different people about the same question, we face the following dilemma: We want them to be free to mention any actor who comes to mind, whoever is involved in, let’s say innovation in the poultry sector in Ethiopia or managing small reservoirs in Ghana. But then we might want to combine these maps to get a master view of the problem. And what seemed like a minor oddity, the fact that people give the same actor different names, or use more or less condensed actor labels (do they just say NGOs or give us the specific names of the NGOs involved, do they name individuals by their name or position?) can become a major problem when trying to combine the maps. Even just minor spelling differences means that we have to fiddle around with the data manually where a click or two should be able to do the job for us.
On the other hand, especially in fields where we are not the experts, it is absolutely unrealistic, that we could come up with a pre-defined list of actors, especially if you want to include formal and informal players. And it is one of the big strengths of this method that it allows you to explore actor constellations even in areas where you don’t even really know what questions to ask.
How do we combine the need to explore and be open with the need for consistency in the labeling of data?
In one of our current projects we will try out an approach that is very remotely inspired by the Delphi approach, we will let our experts build on each others’ knowledge: The first interview will start from a blank slate, asking: Who are all the actors involved?
For the second interview we will write all the actors mentioned by interview partner 1 on prepared stickynotes and ask our respondent to choose whichever they want, plus add any that are missing. Interview 3 will have all the actors from 1 and 2 to choose from, plus any that respondent 3 adds.
I’m curious to see how this goes. Will the later Net-Maps have significantly more actors than the earlier ones? Will we find a good balance between openness and consistency? This approach will work better for some questions than for others and it only really makes sense if your final goal is creating a consensus map of a common network. That means it could work for questions such as: “Who is involved in developing this policy?” but not really for questions like “Who do you personally go to for information about job opportunities?”