I was wondering if there was any python (or R) packages out there that can do spatial clustering on ranked data?

For each US state I spatially varying value for education rankings. For example, for Virginia the ranking may be 50 compared to a ranked value of 10 for the state of Missouri. These are not categorical (e.g. A vs B) or ratio (e.g. 1.5), but ranked in the sense that a value of 50 for Virginia does not mean that this state is 5 times higher in educational value that Missouri. Its synonymous to ranked outcomes at the end of a marathon (e.g. 1st, 2nd,....places). Most of the stuff I've been finding online relate to the relationship between records across different attribute themes in tabular data (e.g. Gower distance). I'm looking for a spatially clustering method such as Moran's I where the attribute vector being analyzed is an education rank value for each state polygon.

  • Sorry, I do not understand: do you have spatially variable data with ranks (a map)? The way you put it seems that Virginia has the same level of 50, is it not a spatial cluster already? – Andrea Massetti Feb 18 '19 at 5:57
  • Hi Andrea, sorry about the misleading information. Your interpretation is correct. Education is spatially variable with ranks. I have a polygon shapefile for the US with each state having a educational ranked value. – Rigid90 Feb 18 '19 at 6:02
  • So you have 1 value for each state? – Andrea Massetti Feb 18 '19 at 6:04
  • Yes, that's correct. – Rigid90 Feb 18 '19 at 6:05
  • What do you want your final map to represent? – Andrea Massetti Feb 18 '19 at 6:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.