While trying to represent my data in the most faithful and useful way, I recently discovered the little free software ScapeToad that makes cartograms.
Very well documented, this software lets users distord their data in function of one parameter, be it population or GDP.. I used the tool to highlight the importance of some alternative open spaces (meaning open spaces that visitors surveyed in the Farlington Marshes or the Milton Common would use if the open spaces in which they were surveyed were to disapear). Trying to make the map of these open spaces lead me to a scale issue: some visitors mentionned open spaces in Portsmouth (so local open spaces) while others said they would go to National Parks, some of them quite far. This meant that representing this on a topographically true map (where distances are kept) probably required something bigger than an A4 map while the size of the alternative open space did not matter as much as other information I could add (number of people who said they would use the open space, or distance from their house to the open space for example). I decided therefore to use a cartogram to represent all the open spaces into one easy to read map. In order to do so, I followed three steps
– first the creation of a layer of centroids of all the open spaces,
– then the creation of Voronoi polygons, so that I had only adjacent polygons,
– and finally using ScapeToad to make cartogram with the ‘number of people who said they would use the space as an alternative open space’ as a parametre
This gave me a map on which the size of the open space depends first on the distance from other open space (as the Voronoi polygons are bigger when centroids are further) and on the potential number of people using these open spaces.