type of variables and the use of Moran's I

PinkOrchid

Hi everyone,

I am wondering if anyone can direct me towards a source where I can find an answer to this question: if the variable I am investigating for spatial autocorrelation is not continuous, is it still appropriate to use Moran's I?

I am referring to binary data (e.g. high versus low access to stores) or multinomial data (e.g., high, medium, and low access to stores). The textbooks I consulted indicate that Moran's I can be used for continuous, ordinal, and interval data. However, all examples concern continuous data. I read a paper looking at the spatial autocorrelation in an obesity variable (non-obese versus obese) for individuals living in the same area. I am not sure that is right, based on what I read.

Can someone help clarify this issue for me?
Thank you very much!

DanPatterson_Retired

There is the Joins Count test...but for the life of me I can't find a decent link other that in old text books