Go back to your initial Knowledge Areas map. You should now be able to cluster the insights you have developed in the newspaper headlines exercise and be able to sort them into each of the areas: Problem, Context, and Past Efforts.

There may be information that does not fit into any of the three areas. If so, it is ok to start a new theme with a cluster. Clusters are where the information you have gathered from your different sources seem to be connected and can be ‘clustered’ into recognisable themes and patterns.

Once you have created this initial clustering, you may want to draw connections across clusters as well. This can help you see the ‘problem space’ from different angles. You can use pens or string to draw the connections. Make sure that you label each connection, so that you don’t end up with an unintelligible meatball-and-spaghetti diagram!

• What patterns start to emerge with respect to your research questions?
• Can you identify common features, sources or elements?
• What are the possible explanations that come out of this exercise?

Consider what your clustering tells you about how the problem is starting to be framed, and whether it would be recognisable to members of the crisis-affected population. Consider also whether your clustering and framing might be overly influenced by the way the humanitarian system and architecture currently approaches this problem.

Further Inspiration

Centre for Care Innovations (2017) Catalyst Method: Affinity Clustering.
Further guidance on clustering